Precision Physiotherapy & Sports Training: Part 1

Precision
Physiotherapy &
Sports Training:
Part 1:
Hardwarelandscape from
computer visiontowearable
sensors, and alight intro for UX
requirements toensure adherence
and engagement.
Version “29/05/2020“
Petteri Teikari, PhD
High-dimensionalNeurology,Queen’sSquareof
Neurology,UCL, London
MSc Electrical Engineering / PhD Neuroscience
https://www.linkedin.com/in/petteriteikari/

Aboutthe Presentation
“Quick” intro for:
●
Physiotherapists, socialworkers, clinicians about the hardware
and/with deep learning.
●
For computer scientists and engineers about clinical
rehabilitation
In order to make cross-disciplinarycommunication “a bit
more effective”andprovide seeds for further self-directed
learning.

Precision
Physiotherapy
Asthe trend istothrow precision prefix
infrontofthe field boosted withfancier
models, oftenincludingdeep learning.
I adoptthe same herefor→ I adopt the same here for
“quantified exercise”, that could be useful
forpost-surgeryphysical rehabilitation (e.g. ACL tear),
post-stroke rehabilitation,elite-level/entry-level
sportsstrengthand conditioning,etc.

YogatrainingwithYogAIanda
RaspberryPismartmirror
https://www.raspberrypi.org/blog/
yoga-training-with-yogai-and-a-ra
spberry-pi-smart-mirror-the-magp
i-issue-80/
1st
orderapproximationof“PrecisionPhysiotherapy”
Quantifyexercise biomechanics throughposeestimation from videofeed (“computer vision2)
http://openaccess.thecvf.com/content_cvpr
_2018/papers/Nie_Human_Pose_Estimatio
n_CVPR_2018_paper.pdf
NationalUniversityofSingapore
https://github.com/NieXC/pytorch-pil
https://www.youtube.com/watch?v=prhGv1Ws2JY
http://groups.inf.ed.ac.uk/calvin/synchronic_activities_stickmen/
WhatKinect gamesare best forexercise?
https://www.quora.com/What-Kinect-games-are-best-for-exercise

PoseEstimation quicktech intro
https://youtu.be/dxOHmvTaCN4

PoseEstimation
(Computer Vision)
Examples
Thatyoucould
wanttodesign

AITrainer for any movementlearning
Wearables,BiomechanicalFeedback,andHumanMotor-Skills’
Learning&Optimization
XiangZhang,GongbingShan,YeWang,BingjunWanandHuaLi
Appl.Sci.2019,9(2),226;https://doi.org/10.3390/app9020226
While real-time physiological and biochemical biofeedback have seen routine
applications, the use of real-time biomechanical feedback in motor learning
and training is still rare. On that account, the paper aims to extract the specific
research areas, such as three-dimensional (3D) motion capture, anthropometry,
biomechanical modeling, sensing technology, and artificial intelligent (AI)/deep
learning, which could contribute to the development of the real-time biomechanical
feedback system. The review summarizes the past and current state of
biomechanicalfeedbackstudiesin sportsandartsperformance
15-segmentbiomechanicalmodelingoftheGrandeJeté(a) inBallet[Shan2005]
andtheAxeKick(b) inTaekwondo[Yuetal.Arch.Budo2012; Citedby15].
The two-chain model of human motor-skills. (a) The possible locations of the six
wearables for human motor-skills’ tracking; (b) A ballet skill; (c) A Indian dance skill;
(d) Baseball pitch; and, (e) Bicycle kick in soccer (the three-dimensional (3D)
motiondatawasgeneratedinShan’sBiomechanicsLab).
The framework can serve as a basis for developing real-time
biomechanical feedback training in practice. In order to creating a
feasible, reliable, and practical biomechanical feedback tool for
athletic and artistic motor-skills’ learning and optimization, the
massive and diverse motor-skill datasets have to be built first. The
big data could be obtained by a synchronized
measurement from 3D motion capture and IMUs. Currently,
gaining high-quality, full-body motion data cross sports and arts
performance would be the vital step for the real-time biomechanical
feed-backdevelopment.

Teaching Motor Skills Drawing JapaneseCharacters
AssistingMovementTrainingandExecutionWithVisualand Haptic
FeedbackRelatedarticles
Marco Ewerton,David Rother,Jakob Weimar, GerritKollegger,Josef Wiemeyer,JanPetersand GuilhermeMaeda
TechnischeUniversität Darmstadt,Max PlanckInstituteforIntelligent System,ATRComputationalNeuroscienceLabs
FrontiersinNeurorobotics,May2018|https://doi.org/10.3389/fnbot.2018.00024
In the practice of motor skills in general, errors in the execution of movements
may go unnoticed when a human instructor is not available. In this case, a
computer system or robotic device able to detect movement errors and propose
corrections would be of great help. This paper addresses the problem of how to
detect such execution errors and how to provide feedback to the human to
correct his/her motor skill using a general, principled methodology based on
imitationlearning.
The core idea is to compare the observed skill with a probabilistic model
learned from expert demonstrations. The intensity of the feedback is
regulated by the likelihood of the model given the observed skill. Based on
demonstrations, our system can, for example, detect errors in the writing of
characterswithmultiplestrokes.
Moreover, by using a haptic device, the HaptionVirtuose6D, we demonstrate a
method to generate haptic feedback based on a distribution over trajectories,
which could be used as an auxiliary means of communication between an
instructor and an apprentice. Additionally, given a performance measurement,
the haptic device can help the human discover and perform better movements to
solve a given task. In this case, the human first tries a few times to solve the
task without assistance. Our framework, in turn, uses a reinforcement
learning algorithm to compute haptic feedback, which guides the human toward
better solutions.
x trajectories of corresponding strokes of multiple
instances of a Japanese character. (A) Before time
alignment. (B) After time alignment using DTW and
our extensiontodealwith multipletrajectories.

VirtualtrainingforMartialArtsandCombatSports
HumanActionsAnalysis:TemplatesGeneration,Matchingand
VisualizationAppliedtoMotionCaptureofHighly-SkilledKarate
AthletesSensors2017,17(11),2590;https://doi.org/10.3390/s17112590
Motionanalysissystemsasoptimizationtrainingtoolsincombat
sportsandmartialartsEwaPolak,JerzyKulasa,AntónioVencesBrito,
MariaAntónioCastro,OrlandoFernandes
http://revpubli.unileon.es/ojs/index.php/artesmarciales/article/view/1687
InertialSensorsforPerformanceAnalysisinCombatSports:A
SystematicReviewSports2019,7(1),28;https://doi.org/10.3390/sports7010028
Inertial sensors are one technology being used for performance monitoring. Within
combat sports, there is an emerging trend to use this type of technology; however, the use
and selection of this technology for combat sports has not been reviewed.A total of 36
records were included for review, demonstrating that inertial measurements were
predominatelyusedfor measuringstrikequality.
Sportsscience-based researchonthesportof muaythai:Areviewof
theliteraturehttp://wjst.wu.ac.th/index.php/wjst/article/view/2243
ConcurrentValidityand
ReliabilityofaLinear
PositionalTransducerand
anAccelerometerto
MeasurePunch
Characteristics
http://doi.org/10.1519/JSC.00000
00000002284
Anaccelerometer (Crossbow)
andalinearpositionaltransducer
(GymAware)wereusedto
examinepeak velocityand
accelerationofeachpunch.Thus,
theGymAwarelinear positional
transducerisanacceptable
measurementtoolforthe
quantificationofpunchspeedfor
straightpunchesinuntrained
adults.

Virtualtraining for Baduanjin
OliveX is a Hong Kong-based company
focused on fitness-related software, serving
more than 2 million users since we first
launched in 2018. Many of our users are
elderly and our Baduanjin app helps
them practice Baduanjin while
minimizing the possibility of injury. To achieve
that, we utilize the latest artificial intelligence
technology in our app to automatically
detect Baduanjin practicing moves and
provide corresponding feedback to our
users.
By using the “Smart Baduanjin” app, users can determine if they are performing the moves correctly by using AI
to track their movements. By leveraging the latest machine learning technology, we hope to replace the traditional learning
approach in which users simply follow an exercise video with a more enjoyable interactive experience in which
users get feedback on their body movements in real time. We also hope that these features could help the elderly
topracticeBaduanjinmoreeffectivelyandreducetherisk ofinjury.
Challenges on mobile devices After finishing the deep
learning model, our next step was to deploy our models on iOS
and Android mobile devices. At first, we tried TensorFlow Mobile.
But since we needed to get recognition results in real time,
TensorFlow Mobile was not a viable option since its performance
did not meet this requirement. As we were trying to solve the
performance challenge, Google released TensorFlow Lite, which
wasabigleap fromTensorFlowMobilein termsof performance.

Virtualtraining for Dancing
SmartTechnologyforSupportingDanceEducation
AugustoDiasPereiradosSantosTheUniversity ofSydney
UMAP'17
https://doi.org/10.1145/3079628.3079709
My aim is to design, implement and evaluate a conceptual and technological solution that
captures students' movement using wearable devices and help dance
teachers and students enhance their awareness and promote reflection
regarding dance skills acquisition using automated personalised feedback (charts,
tables,text,etc.).
I will explore how to acquire movement data that can represent key aspects of social
dance learning, and how to use these data to support of students and teachers. For
this, I created a mobile app that records students' movement while they are practicing
danceexercisesandcreatesadancelearnermodel.
The learner model's features are exposed through the Open Learner Model to
students and their teachers in order to support reflection and increase awareness. With
the proposed work I expect to generate a deeper understanding of the aspects of the
dance learner model which can be used to promote personalization and adaptation,
andpositivelyimpactdancelearning.
HappyFeet:RecognizingandAssessingDanceontheFloor
AbuZaherMdFaridee,SreenivasanRamasamyRamamurthy,HMSajjad
Hossain,NirmalyaRoy University ofMaryland
HotMobile'18
https://doi.org/10.1145/3177102.3177116
Recognizing dance steps with fine granularity using wearables is one of those
exciting applications. In a typical dance classroom scenario where the instructors are
frequently outnumbered by the students, accelerometer sensors can be utilized to
automatically compare the performance of the dancers and provide informative
feedbacktoallthestakeholders,forexample,theinstructorsandthelearners.
However, owing to the complexity of the movement kinematics of human
body, building a sufficiently accurate and reliable system can be a daunting
task. Utilization of multiple sensors can help improve the reliability, however most
wearable sensors do not boast sufficient resolution for such tasks and often
sufferfromvarious datasampling,deviceheterogeneity and instability issues.
To address these challenges, we introduce HappyFeet, a convolutional neural
network based deep, self-evolving feature learning model that accurately
recognizes the micro steps of various dance activities (Indian classical) performed by
aprofessionaldancer.

Virtualtraining for Yoga
Validityofalow-costwearabledeviceforbodyswayparameter
evaluationsA.Rouis,N.Rezzoug &P.GorceToulon,HandiBio
ComputerMethodsinBiomechanicsandBiomedicalEngineering
Volume17,2014http://dx.doi.org/10.1080/10255842.2014.931671
Datawererecordedwitha10bits,low-power,three-axialaccelerometer
MMA8453Q andaforceplatformAMTI’sAccuSwayPLUS
at50Hz.
ThesubjectswereaskedtoexecuteoneyogaexercisenamedTadasana.
Itisdecomposedinthreestaticphases.Duringthefirstphase,thesubject
standsinthestandardpositionwitharmslyingalongsidethebody;inthe
secondphase,bothupper limbsareraisedhorizontallyinthefrontalplane;
andinthethirdphasetheupper limbsareraisedverticallyabovethehead.
Thesubjectsexecutedthethreeposturesinarowand30sofsteady
statewereextractedfromeachphase
https://doi.org/10.1007/s11042-018-5721-2 (2018): “In this paper, we propose a yoga
self-training system, which aims at instructing the practitioner to perform yoga poses
correctly, assisting in rectifying poor postures, and preventing injury. Integrating computer
vision (OpenCV) techniques, the proposed system analyzes the practitioner’s posture
from both front and side views by extracting the body contour, skeleton, dominant axes, and
feature points. Then, based on the domain knowledge of yoga training, visualized
instructions for posture rectification are presented so that the practitioner can easily
understand how to adjust his/her posture”

Virtualtraining for Yoga for low-vision/blind
DesignandReal-WorldEvaluationofEyes-FreeYoga:An
ExergameforBlindandLow-VisionExercise
Kyle Rector, Roger Vilardaga, LeoLansky, Kellie Lu, CynthiaL. Bennett, Richard E.
Ladner, and Julie A. Kientz Department ofComputer Science, University ofIowa
ACMTransAccessComput.2017Apr;9(4): 12.doi: 10.1145/3022729
People who are blind or low vision may have a harder time
participating in exercise due to inaccessibility or lack of
encouragement. To address this, we developed Eyes-Free Yoga
using the MicrosoftKinect (withKinectfor Windows Toolkit,Python,NonVisualDesktopAccess
(NVDA) screen reader on the laptop)
that acts as a yoga instructor and has
personalized auditory feedback based on skeletal tracking.
We conducted two different studieson two different versions of Eyes-
Free Yoga: (1) a controlled study with 16 people who are blind or low
vision to evaluate the feasibility of a proof-of-concept and (2) an 8-
week in-home deployment study with 4 people who are blind or low
vision, with a fully functioning exergame containing four full workouts
and motivational techniques. We found that participants preferred
the personalized feedback for yoga postures during the
laboratory study. Therefore, the personalized feedback was used as a
means to build the core components of the system used in the
deploymentstudyandwasincludedinbothstudyconditions
On average, motivational techniques increased participant’s
user experience and their frequency and exercise time. The
findingsofthiswork have implicationsfor eyes-free exergame design,
including engaging domain experts, piloting with
inexperienced users, using musical metaphors, and designing for
in-homeusecases.
https://youtu.be/cm_ghJPqj70
https://vimeo.com/76583949

Virtualtraining forFencing
Automaticanalysisof techniquesandbodymotionpatterns
insport.PhDthesisbyFilipMalawski
https://www.linkedin.com/in/filip-malawski-80228a6/
"This would not only provide a useful tool for fencing footwork training, but also
allow to easily gather more data which could be used for further development of
action analysis methods. The detection of other actions and the analysis of their
performance would be interesting. It is worth noting, that joint research in this
area is currently being conducted with the Delta Fencing Center (
http://www.deltafencingcenter.com/)
, located in California, USA ""Another interesting manner of
providing feedback would be virtual reality (VR). By tracking the athletes’motion,
virtual exercises could be possible, maybe even including virtual opponents,
controlled by artificial intelligence algorithms. The visual cues and feedback for
bladework practice could be presented by using VR as well. The main
advantage in this case would be a much lower cost - while AR glasses are
expensive, VR can be achieved with a simple low-cost cardboard adapter used
with a smartphone. ""Finally, it would be beneficial to adapt the results of this
research to other sports disciplines. Since similar problems occur in other
sports as well, it should be possible to develop dedicated motion analysis
methodsbased on theonesproposedin thiswork."

Virtualtraining forPoleDancing/AerialCircus/Stretching /Calisthenics
endlessopportunities
StretchIt - Stretching and Flexibility Videos
https://play.google.com/store/apps/details?id=com.stretchita
pp.stretchit&hl=en
https://youtu.be/RHQF65IzscM
https://youtu.be/YELQ2Yeh19s
https://youtu.be/PcMihvlaFPc
https://youtu.be/xJwwioOcE4E

Rock Climbing andBouldering Assistant
AutomatedRecognitionandDifficultyAssessmentofBoulder
Routes
AndréEbert,KyrillSchmid,ChadlyMarouane,ClaudiaLinnhoff-Popien
HealthyIoT2017: InternetofThings(IoT) Technologiesfor HealthCare
https://doi.org/10.1007/978-3-319-76213-5_9
Incontrasttotheassessmentofrecurrenthumanmotionasproposedin
climbingactivitiesmaynotbedescribedbyfeatureslikesimilarity,
periodicity,or runtime.Onereasonfor thedifficultyofusingtemporal
featuresisthatdifferentboulder andclimbingroutesareofsignificantly
varyinglengthsandconsistencies.Togetherwithskill-dependentascent
times,thatmakesithardtofindgeneralizable,time-dependent
featuresforawholeclimbingactivity.
Thelackofperiodicityresultsinthefact,thatcomparisontoqualitatively
labeledpatternsisalsonotfeasible,e.g.,apushupofgoodqualityvs.one
ofbadquality.Toovercomethoseissues,weuse someassumptions
basedonclimbingtheory:anincreasedlevelofroutedifficultyis
indicatedbyinaccurategrippingandincreaseduseofstrength
duringtransitionperiods,whilea tremblingoftheclimber’slimbs
occursmoreoften withinrestperiodsbecauseofexhaustionand
imperfectcontrol.Thecoreskillscontrol,stability,speed,andeconomical
useofstrengthareharder toachievefor difficultroutesandtherefore
seemsuitableasatheoreticalbasisforfeatureengineering.
WeappliedtheSensXsensorarchitecture https://arxiv.org/abs/1703.02847
asa
technicalbasis. Thefour externalmBientLabsensorplatforms(rightarm,
leftarm,rightleg,leftleg) providesampleratesofroughly40Hzwhilethe
processingunit(chest) provides50Hzfor accelerationand100Hzfor
rotationdata.AlldevicesareconnectedbyBluetoothLowEnergy(BLE)
andaresynchronizedbytheprocessingunit.

HomeExercise with Depth Cameras (RGB-D)
GymCam:Detecting,Recognizingand
TrackingSimultaneousExercisesin
UnconstrainedScenes
IlktanAr; YusufSinanAkgul GebzeInstituteofTechnology
IEEETransactionsonNeuralSystemsand
RehabilitationEngineering( Nov.2014 )
https://doi.org/10.1109/TNSRE.2014.2326254
Computerized recognition of the
home based physiotherapy
exercises has many benefits and it
has attracted considerable interest
among the computer vision
community. However, most
methods in the literature view this
task as a special case of motion
recognition. In contrast, we propose
to employ the three main
components of a physiotherapy
exercise (the motion patterns, the
stance knowledge, and the
exercise object) as different
recognition tasks and embed them
separately into the recognition
system.

HomeExercise with Smart Speakers
Amazon Echo, GoogleHomewith Alexa, Cortana, Google Assistant, Siri and whatever you havespying you
Home-basedexercisesystemforpatientsusingIoT enabled smart
speakerJayneelVora ; Sudeep Tanwar; SudhanshuTyagi; NeerajKumar; JoelJ P CRodrigues(2017)
https://doi.org/10.1109/HealthCom.2017.8210826 -Citedby9 -Relatedarticles
There is no cost effective and non-
complex methods available to
quantify the exercises performed
by the patient. In this paper, a study
was performed to check the validity
and efficiency of a system
consisting of a Smart IoT
enabled speaker, which contains
an orchestrator. Which is speech
learning unit, an exercise
database at the edge, and
connected to the cloud, where the
generated reports are stored and
transferred for further analysis, if
required. We report the efficiency of
the system compared to the ratings
of a physical therapist, a
standard currently being used.

Whatdoyoudo withthis quantifiedexercise?
In sports medicine, it is of an interest to be able to predict the injury probablity, and reduce the stress on that athelete
when being close ofbeing “statistically injured” RGBComputer vision alonenotreally enough?
A 32-year-old professional football player is sprinting
towards the goal. He feels sudden pain in his right
hamstring, falls to the ground and cannot continue.
Medical assessment reveals a torn right semimembranosus
and he will miss the rest of the season. The medical staff
might ask themselves: ‘‘Could our screening processes have
identified possible causal factors or maybe predicted this
injury? Could we have prevented it?’’
In elite professional team sports such as football,
preventing an injury is big business Jones et al. 2019
. For
every player missing through injury the cost to an elite
football team is approximately €20,000 (US$24,000) per
day [Jan Ekstrand 2016, UEFA Injury Study Lead Expert,
Linköping, Sweden]
ManchesterUnitedFootball Club, ArthritisResearchUKCentrefor Epidemiology, Centre
for Biostatistics, Universityof Manchester, Centrefor PrognosisResearch,Departmentof
Health Professions
https://dx.doi.org/10.1007%2Fs40279-018-0928-y
http://www.thermohuman.com/thermography-
application-in-sports-and-physiotherapy/

Even thenon-elitestrength athleteswouldbenefitfrom“injurymonitoring”
EpidemiologyandTrendsofWeightlifting-RelatedSprainsandStrainsthat
PresentedtoEmergencyDepartmentsintheUnitedStates
StevenA.Burekhovichetal.
DepartmentofOrthopaedicSurgery andRehabilitation Medicine,StateUniversity ofNewYork(SUNY),DownstateMedicalCenter,Brooklyn,NY Departmentof
OrthopaedicSurgery,Johns Hopkins University,Baltimore,MD
Journal of Long-Term EffectsofMedical Implants>Volume 28, 2018 Issue2
https://doi.org/10.1615/JLongTermEffMedImplants.2018026168
Despite potential health benefits of weightlifting and physical activity, individuals can suffer from
anumberofmusculoskeletalinjuries.Thisstudyaimedto:
●
Compare incidence and annual trends of different weightlifting injury types presenting to
emergencydepartmentsintheUnitedStatesand
●
Identify frequency and annual trends of weightlifting-related sprains and strains to each
bodypart.
The National Electronic Injury Surveillance System was queried to identify all weightlifting-
related injuries from 2010–2016. Incidence and annual trends of various types of
weightlifting-related injuries were compared during the study period. Furthermore, frequency
and annual trends of weightlifting-related sprains and strains to different body parts were
assessed.
The weighted estimated annual incidence of weightlifting-related injuries significantly
increased from 86,910 in 2010 to 109,961 in 2016. The most common weightlifting-
related sprains and strains involved the lower trunk (29.4%), shoulder (22.6%), upper trunk
(17.3%),neck(6.5%),upper arm(5.6%),wrist(4.8%),knee(3.4%),andelbow(2.6%).
There was a significant increase in the frequency and trends of sprains and strains that
involved the lower trunk. Weightlifting-related injuries have increased, of which sprains and
strains were the most common. Additionally, the most commonly affected body part was the
lower trunk. Further studies are needed to determine the etiologies of weightlifting-
related lower trunk sprains/strains. This study may be beneficial to weightlifters,
highlighting commoninjurytypes,therebyallowing themtotake preventativemeasures.
Incidenceandcharacteristicsofacuteandoveruseinjuriesin
elitepowerlifters
ThomasReichel,MartinMitnacht,AnnabelFenwick,Rainer Meffert,
OlafHoos&KaiFehske
DepartmentofOrthopaedicTrauma,Hand,PlasticandReconstructiveSurgery,University HospitalWuerzburg,
Cogent Medicine2019
http://doi.org/10.1080/2331205X.2019.1588192
In this study, we were able to gain new insights into the influence of
powerlifting equipment, preventive and regenerative methods as well
as training periodization on the rate of acute and overuse injuries in
powerlifting. Future studies should develop an optimized set of
preventive exercises and lifestyle recommendations individualized
to the relevant preconditions and risk factors of each athlete to reduce
orpreventacuteandoveruseinjuries

Specific‘smartsensingclothing’ requires extramotivation fromthe end-user,
https://www.sporttechie.com/smart-apparel-connected-coaching-asensei-tr
x-yoga-rowing-athos
https://www.fastcompany.com/90458891/the-next-big-thing-in-sports-cloth
es-that-give-you-perfect-form
serving elite athletes and the military in 2017"
https://www.youtube.com/watch?v=EBnK6i0zsnU

notjustsports Clinicalphysiotherapyforrehabilitation
Mostoftherehabisboring,andyouliketovisualizethetherapyprogress,withorwithouttrendy buzzwordey gamification.
Intheend,mostofthepatientsare nottech-savvy,andsufferfromsocialisolation,andinworstcasesdepression.
Opportunitiesofa MachineLearning-basedDecisionSupport
System for StrokeRehabilitationAssessment Min HunLee, Daniel
P.Siewiorek,AsimSmailagic,AlexandreBernardino,Sergi Bermúdez i
Badia(Submittedon 27 Feb2020(v1), last revised2Mar2020(thisversion,
v2)) https://arxiv.org/abs/2002.12261

A fieldof “Seriousgames” existe.g.for clinicalapplications
SeriousPlayConferenceisaleadershipconferencefor boththose
whocreateseriousgames/simulationsandthosewhoimplement
game-basedlearningprograms.
https://seriousplay-montreal.com/
InternationalConferenceonSeriousGamesandApplicationsfor
Health,IEEE SeGAH2019http://www.segah.org/2019/

Clinicalgames vsMainstreamGames
Clinical Rehabilitation ExperienceUtilizing SeriousGames: Rehabilitation
Technology and a TechnicalConceptfor Health Data Collection
byMichael Lawo (Editor),Peter Knackfuß(Editor)
http://doi.org/10.1007/978-3-658-21957-4
Makesurethatyour game hasclinical
value, butitisnottoo boringkeeping
patientsengaged

“Seriousgames” needclinicalvalidationandRCTs*RCT randomized clinical trials
https://doi.org/10.1007/978-3-319-66122-3_1
Clinicianperceptions
ofaprototype
wearableexercise
biofeedbacksystem
fororthopaedic
rehabilitation:a
qualitative
exploration
RobArgent,Patrick
Slevin, Antonio
Bevilacqua,Maurice
Neligan, AilishDaly,
BrianCaulfield
BMJOpen
2018;8:e026326.
http://dx.doi.org/10.1136/
bmjopen-2018-026326
Citedby2
Relatedarticles

TechorHumanfirst
Shouldnotbemutually
exclusive,anddonotdo
”engineeringtech”,i.e.
techfortechjustbecause
youthinkitiscoolwhenno
onewantstouseit

Technology-firstapproach forengagement#1A
Multimodaladaptiveinterfacesfor3Drobot-
mediatedupperlimb neuro-rehabilitation:An
overview ofbio-cooperativesystems
DavideSimonetti,LoredanaZollo,EugeniaPapaleo,Giorgio
Carpino,Eugenio Guglielmelli
RoboticsandAutonomousSystems
Volume 85, November2016,Pages62-72
https://doi.org/10.1016/j.robot.2016.08.012
Citedby11
Robot-mediated neuro-rehabilitation has been proved to be an effective therapeutic approach for upper limb motor
recovery after stroke, though its actual potential when compared to other conventionalapproaches has still to be fully
demonstrated. Most of the proposed solutions use a planar workspace. One key aspect for influencing motor recovery
mechanisms, such as neuroplasticity and the level of motivation and involvement of the patient in the exercise, is
the design of patient-tailored protocols and on-line adaptation of the assistance provided by the robotic agent to
the patient performance. Also, when abilities for performing activities of daily living shall be targeted, exercises in 3D
workspaceare highly preferable.

Technology-firstapproach forengagement#1B
Notexactly (yet) themostcommon athomerehabilitationmethodto
haverobot-assisted/ exoskeleton -basedexercises
WenhaoDeng et al. (2018)
https://doi.org/10.1109/RBME.2018.2830805

Technology-firstapproach forengagement#1C
Long-TermTrainingwithaBrain-MachineInterface-BasedGait
ProtocolInducesPartialNeurological Recoveryin Paraplegic
Patients AnaR.C.Donati etal.
Neurorehabilitation Laboratory,Associação AlbertoSantosDumont paraApoioà Pesquisa(AASDAP),Sâo Paulo,BrazilEdmondandLily SafraInternational InstituteofNeuroscience,SantosDumont Institute,Macaiba,Brazil /DukeUniversity
ScientificReportsvolume6,Articlenumber:30383(2016)
https://doi.org/10.1038/srep30383 | Cited by140 -Relatedarticles
CombinedrTMSandvirtual
reality brain–computer
interfacetrainingformotor
recovery afterstroke
NN Johnsonetal.(2018)
Department ofBiomedicalEngineering,UniversityofMinnesota
J.NeuralEng.15016009
https://doi.org/10.1088/1741-2552/aa8ce3
Combining repetitive transcranial
magnetic stimulation (rTMS) with
brain–computer interface (BCI)
training can address motor
impairment after stroke by down-
regulating exaggerated inhibition from
the contralesional hemisphere and
encouraging ipsilesional activation.
The objective was to evaluate the
efficacy of combined rTMS + BCI,
compared to sham rTMS + BCI, on
motor recovery after stroke in
subjectswithlastingmotorparesis.

Technology-firstapproach forengagement#2
AdvancesinAutomationTechnologiesfor
LowerExtremityNeurorehabilitation:A
ReviewandFutureChallenges
WenhaoDeng et al. (2018)
IEEE Reviewsin Biomedical Engineering( Volume:11)
https://doi.org/10.1109/RBME.2018.2830805
“This survey paper provides a
comprehensive review on recent
technological advances in wearable
sensors, biofeedback devices, and
assistive robots. Empowered by the
emerging networking and computing
technologies in the big data era, these
devices are being interconnected into
smart and connected rehabilitation
systems to provide nonintrusive and
continuous monitoring of physical and
neurological conditions of the patients,
perform complex gait analysis and
diagnosis, and allow real-time decision
making, biofeedback, and control of
assistive robots.”

DeepLearning for MusculoskeletalPhysiotherapy
Artificialintelligenceandmachinelearning|applicationsin
musculoskeletalphysiotherapy
Musculoskeletal Science and Practice, Volume 39, February 2019
ChristopherTack, Guy'sand St Thomas' NHSFoundation Trust,Guy's Hospital,Great Maze Pond,SE1 9RT, London,
UK
https://doi.org/10.1016/j.msksp.2018.11.012
This review outlines key applications of supervised and unsupervised
machine learning in musculoskeletal medicine; such as diagnostic
imaging,patientmeasurementdata,andclinicaldecisionsupport.
Potential is apparent for intelligent machines to enhance various areas of
physiotherapy practice through automization of tasks which involve data
analysis, classification and prediction. Changes to service provision through
applications of ML, should encourage physiotherapists to increase their
awareness of and experiences with emerging technologies. Data literacy
should be a component of professional development plans to
assist physiotherapists in the application of ML and the preparation of
informationtechnologysystemstousethesetechniques.
Clinicaldecisionsupportsystems (CDSS) provide recommendations on
diagnosis and treatment (Musenetal.,2014). Systems have been
developed for LBP: for example the StartBack riskstratification tool
which identifies prognostic indicators to classify individuals into
riskgroups (Hilletal.,2008). Nijeweme-d'Hollosyetal.(2016) developed a
digital CDSS to stratify patients to self-management, GP attendance or
self-referral to physiotherapy. An ontology and decisiontree to classify
subjects was developed according to 43 decision factors; such as
general factors (e.g. occupation), ‘psychosomatic’ factors (e.g. depression,
kinesiophobia);andseriouspathologysigns(i.e.redflags).
Recentdevelopmentsinhumangaitresearch:parameters,approaches,
applications,machinelearningtechniques,datasetsandchallenges
Artificial Intelligence Review January2018
ChandraPrakash, RajeshKumar and NamitaMittalMalaviya National Institute ofTechnologyJaipurIndia
https://doi.org/10.1007/s10462-016-9514-6
Human gait provides a way of locomotion by combined efforts of the brain, nerves,
and muscles. Conventionally, the human gait has been considered subjectively
through visual observations but now with advanced technology, human gait analysis
can be done objectively and empirically for the better quality of life. In this paper, the
literature of the past survey on gait analysis has been discussed. This is followed
by discussion on gait analysis methods. Computer vision -based human
motion analysis has the potential to provide an inexpensive, non-obtrusive
solutionfor theestimationofbodyposes.

Social-firstapproach forengagementandmotivation
Manypost-stroke patientsfeelisolated and become depressed leading tosuboptimal
therapyoutcomes

Howself-trackingbiometricsinfluencepatients,medicine
andsociety Formany,self-monitoringis becominganew philosophyforlife,arguesdigital health
journalist andbloggerArturOlesch.
https://www.mobihealthnews.com/content/europe/opinion-how-self-tracking-biometrics-influence-patients-medicine-and-society
For many, self-monitoring is becoming a new philosophy for
life: tech companies and innovators promise a healthier, longer and
better life, with rationalisation and control of every aspect of life
instead of uncertainty. Silicon Valley startups are racing to create a
new “medical Tricorder”, a universal, portable scanning device
for self-diagnosis within a few seconds. Body hacking includes
consumer genomics, DNA-sequencing to define ancestry, and
understandingthe metabolism orgenetichealthrisks.
From the patient's perspective, wearables are not just gadgets
but tools that offer real help. Aron Anderson, who after surviving
cancersurgeryat the age of eight,wasconfrontedwithspending the
rest of his life in a wheelchair. Although medicine was able to cure
him, it did not make him healthy. Wearables helped him to regain
some control over his own health: "I believe that self-tracking and
quantifying is a great tool that has the potential to change a lot of
people’s lives in the future,” says Aron. Over the last few years he has
been doing a lot of self-experimentation and tracking, and the
most useful metrics that he has been able to track are HRV (heart
rate variability), DNA-testingand bio-feedback meditation.
However, digital health technology, including wearables, is not a silver
bullet. It generates opportunities, but also new challenges and
threats.
“In some instances, the movement has be one of obstructions and
complications. From cost to clinical utility, the quantified-self movement
has taken a path with several significant (and valuable) outcomes. In
essence, it has arrived as an option verses an imperative. From a clinical
perspective, care providers see much "consumer data” as
unnecessary and as something that adds ambiguity and complexity to
an already difficult process.
"Things like consumer genomics, heart rate variability, gut flora are still
very much part of the "noise" of new found technology,” comments
Nosta. For the founder of NOSTALAB, the digital health movement is
impacting medicine in important and positive ways. From driving a more
proactive consumer posture around wellness to early disease detection
and prevention, the quantified-self is establishing a "new normal"
in care. Additionally, the shift away from traditional brick and mortar
clinical settings to telemedicine and digital health tools is beginning
to establishpowerfulcost-savings.

CombineTech+(Virtual)HumanConnection
forbestoutcomesandrehabilitationadherence
PhysicalRehabilitation
Examinationof Function
DavidA.Scalzitti
https://fadavispt.mhmedical.com/content.aspx?bookid=1895&s
ectionid=136486692
Promoting Optimal PhysicalExerciseforLife(PROPEL):
aerobic exerciseandself-managementearlyafter stroketo
increasedailyphysical activity
http://dx.doi.org/10.1136/bmjopen-2017-015843
A systematicreview ofmeasuresofadherence to
physical exerciserecommendationsin people with stroke
TaminaLevy, Kate Laver, Maggie Killington, NatashaLannin, Maria Crotty
https://doi.org/10.1177%2F0269215518811903

Futureofdigitalhealthinthefieldofbehavioralmedicine
Thehistoryandfutureofdigitalhealthin
thefieldofbehavioralmedicine
Danielle Arigo, DanielleE. Jake-Schoffman, Kathleen Wolin,
EllenBeckjord, Eric B. Hekler, Sherry L. Pagoto
Eric B.Heklerservesasscientificadvisorto OmadaHealth,ProofPilot,andeEcoSphere.SherryL.Pagotoservesas
scientificadvisertoFitbit.
Journal ofBehavioral Medicine (2019)
https://doi.org/10.1007/s10865-018-9966-z
Here, we highlight key areas of opportunity
and recommend next steps to further advance
intervention development, evaluation, and
commercialization with a focus on three
technologies: mobile applications (apps),
social media,andwearabledevices.
Ultimately, we argue that future of digital health
behavioral science research lies in finding ways
to advance more robust academic-
industry partnerships. These include
academics consciously working towards
preparing and training the work force of the
twenty first century for digital health, actively
working towards advancing methods that can
balance the needs for efficiency in industry with
the desire for rigor and reproducibility in
academia, and the need to advance common
practices and procedures that support more
ethical practices for promoting healthy
behavior.
Althoughitmayseemthat thefieldof
behavioralmedicineisnewto
technology,wehavealong historyof
embracing newtechnologiesin the
pursuitoffosteringbetterhealth
outcomesthroughbehaviorchange.
Thenewest permutationof
digital healthisestablishingnew
opportunitiesfordeveloping
scalableeffectiveinterventions,
butmyriadchallengesremain related
toaligningincentives,methods,
andethicalstandardsbetween
thefieldofbehavioralmedicineand
industrypartnerswhocan facilitate
thescaling.
However, an emergence of
academics is producing and
evaluating tools and resources that
are used in the real world, just as
an emergence of industry partners is
interested in using data and evidence
to create tools that produce the
results they are designed to produce.
The profound risk to the behavioral
science community is in not acting
and finding ways to support the
emerging industry that shares our
values and goals of better health
throughscientificallygroundedwork.

Anddonotforgettheneuroscienceofrehabilitation
Rehabilitativedevicesforatop-
downapproach
GiovanniMorone, Grazia FernandaSpitoni, Daniela De Bartolo, Sheida Ghanbari
Ghooshchy, Fulvia DiIulio, Stefano Paolucci, PierluigiZoccolotti& Marco Iosa
Expert Review of Medical DevicesVolume 16, 2019
https://doi.org/10.1080/17434440.2019.1574567
In recent years, neurorehabilitation has moved
from a ‘bottom-up’ to a ‘top down’ approach. This
change has also involved the technological devices
developed for motor and cognitive rehabilitation. It
implies that during a task or during therapeutic
exercises, new ‘top-down’ approaches are being
used to stimulate the brain in a more direct way to
elicit plasticity-mediated motor re-learning. This is
opposed to ‘Bottom up’approaches, which actat the
physical level and attempt to bring about changes at
thelevelofthecentralneuralsystem.
In the present unsystematic review, we present the
most promising innovative technological
devices that can effectively support rehabilitation
based on a top-down approach, according to the most
recentneuroscientificandneurocognitivefindings.
In particular, we explore if and how the use of new
technological devices comprising serious exergames,
virtual reality, robots, brain computer interfaces,
rhythmic music and biofeedback devices might
provideatop-downbasedapproach.

High-LevelApproach
forthetechpartofyourstudy
design,startupidea
UX
e.g.implementin
VirtualReality

VirtualReality engagementideas Avatar fromphysicalpatient

VirtualRealityengagementideas Turning/OmnidirectionalTreadmills
https://arstechnica.com/gadgets/2018/11/forget-vr-t
readmills-google-patents-motorized-omnidirectional
-vr-sneakers/
●
VirtuixOmni $699
●
CyberithVirtualizer
●
KatWalkKickstarter
●
SpacewalkerVR
●
Infinadeck
https://filmora.wondershare.com/virtual-reality/top-vr-t
readmills.html
Experiences oftreadmill walkingwithnon-immersive virtualreality
afterstrokeoracquiredbraininjury–Aqualitative study (2018)
KarinTörnbom,AnnaDanielsson https://doi.org/10.1371/journal.pone.0209214
Patients’andHealthProfessionals’ExperiencesofUsingVirtual
RealityTechnologyforUpperLimb TrainingafterStroke:AQualitative
Substudy (2018)
HannePallesen,MetteBrændstrupAndersen, GunhildMoHansen,CamillaBieringLundquist,
andIrisBrunner https://doi.org/10.1155/2018/4318678
Gait TrainingafterStroke onaSelf-PacedTreadmill with and without
VirtualEnvironment Scenarios:AProof-of-PrincipleStudy (2018)
CarolL.Richards, AnoukLamontagne, BradfordJ.McFadyen, FrancineDumas, François
Comeau,Nancy-MichelleRobitaille,JoyceFung https://doi.org/10.3138/ptc.2016-97
Combiningthe benefitsoftele-rehabilitationandvirtualreality-based
balancetraining:asystematic reviewonfeasibilityand
effectivenessy (2019)
JonasSchröder,TamayavanCriekinge, ElissaEmbrechts,XantheCelis,Jolien VanSchuppen,
Steven Truijen &WimSaeys https://doi.org/10.1080/17483107.2018.1503738
“VR-based interventions are game-like and therefore seem to provide a
motivational environment which allows longer exercise sessions and greater
adherence to therapy.”

Gym in VirtualReality Overview
VirtualFitness:ReshapingExercise
RichardJ.Èlmoyan KnoxlabsMixedRealityLabaratores
Apr272019
https://medium.com/knoxlabs-vr/virtual-fitness-reshaping-exercise-a03d75c9f3e3
According to the VirtualRealityInstituteof HealthandExercise, statistics show
that since 2016, virtual reality games such as Audioshield have helped burn at least 160
million calories. Universities have quickly jumped to learn more about this concept, and
as the evidence and research compiles, institutions such like San Francisco State
University apply VR to wellness centers and exercise programs to track the virtual
healthbenefitsthattranslatetotherealworld.
We have consistentfitnessprogramssuchasJakePhillips’ 90-DayFitnessChallengeon
the KATWalk TreadmillSystem that exemplifies the possibility of a routine workout
based around virtual reality video-gaming. Which in return questions and redefines
conventionalexerciseasweknowit
In 2018, San Francisco State University’s Kinesiology
Department kick-started a fitness program for
students and staff, incorporating virtual reality
applications to monitor heart rate levels, intake of
oxygen, and other health indicators. The purpose of
this research campaign is to gather data and
statistics, find context within the research, and
furtherelaborateontheexactbenefitsof virtualreality.
https://youtu.be/_TTV5lHpcOo #VirtualReality #SFSUhttp://katvr.com/product/kat-walk/

VR inSportsPsychologyand InjuryRehabilitation
Theuseofvirtualrealityhead-
mounteddisplayswithinapplied sport
psychology
Jonathan M. Bird DepartmentofLifeSciences,BrunelUniversity London,
London,UK
https://doi.org/10.1080/21520704.2018.1563573
This article provides the reader with an
understanding of key components and
concepts associated with VR head-mounted
displays (HMDs). Subsequently, a range of
possible applications within applied sport
psychology are discussed, such as the
training of perceptual-cognitive skills,
relaxation strategies, and injury rehabilitation.
Thereafter, the practicalities of using VR
HMDs are outlined, and recommendations
are provided to applied sport psychology
practitioners wishing to embed this
technologywithintheirpractice.
During rehabilitation, VR environments that simulate training drills can be developed so that
injured athletes can begin training with reduced risk of physical injury. A benefit of using VR
environments in this manner concerns the potential to gamify elements of the rehabilitation
process. Hence, an injured athlete might perform a set of rehabilitation exercises administered
through a VR HMD and have the VR system record an objective measure of success (e.g.,
completion time). A personal leader board might be used, which could reinforce feelings of
progression toward the athlete’s rehabilitation program. Readers are referred to a video illustrating
how the company Rezzil (MiHiepa Sports before) are currently using VR HMDs to assist the
rehabilitation of soccer players in the United Kingdom (VRFocus 2018, May 28 Train and rehabilitate athletes in VR)
Perhaps the most recognizable company currently using VR HMDs to train athletes’
perceptual-cognitive skills is STRIVR. Derek Belch, the founder of STRIVR, recognized that
the typical eye-in-the-sky video footage used to review football plays wasn’t fully representative
of the vantage point experienced by athletes in the competitive arena. Subsequently, STRIVR
recorded 360° videos of specific plays being executed from the perspective of a quarterback.
Thereafter, the athletes could use a VR HMD to review the footage, allowing them to scan the
field of play, anticipate the pass rush, and to identify their receivers. It has been reported that
quarterback Case Keenum watched over 2,500 plays using a VR HMD during his 2017
season with the Minnesota Vikings (ESPN). However, players from other positions can use VR
HMDs to study blitz pickups and moves at the line of scrimmage

VR partof ExerciseImmersion
Ready exerciser one:examiningthe
efficacy of immersivetechnologiesinthe
exercisedomain
Jonathan M. Bird DepartmentofLifeSciences,BrunelUniversity London, London,UK
Doctoral Thesis, Brunel University
http://bura.brunel.ac.uk/handle/2438/18291
The present programme of research
sought to examine the effects of audio-
visual stimuli during exercise, using
immersive, commercially
available technologies. Three
original studies were conducted using
a range of settings (i.e., real-world,
laboratory), methodologies (i.e.,
qualitative and quantitative), exercise
modalities (i.e., gym workouts, cycle
ergometry) and consumer products
(e.g., music-video channels, virtual
reality head-mounted displays) in
order to explore the main research
questionfromvariousperspectives.

Gym in VirtualReality with “IoT Sensors”
WhenVirtualRealityMeetsInternetofThingsintheGym:
EnablingImmersiveInteractiveMachineExercises
FazlayRabbi, TaiwooPark,BiyiFang,MiZhang,YoungkiLee(2018)
MichiganStateUniversity/SingaporeManagementUniversity
https://doi.org/10.1145/3214281
Toward this vision, we present JARVIS, a virtual exercise assistant that is
able toprovidean immersive andinteractivegymexercise experience
to a user. JARVIS is enabled by the synergy between Internet of Things
(IoT) and immersive VR. JARVIS employs miniature IoT sensing devices
removably attachable to exercise machines to track a multitude of
exerciseinformation including exercise types, repetition counts,and progress
withineachrepetitioninrealtime.
Based on the tracked exercise information, JARVIS shows the user the
proper way of doing the exercise in the virtual exercise environment,
thereby helping the user to better focus on the target muscle group. This
machine-attachable approach not only equips exercise machines with
sensing capabilities without being instrumented but also turns JARVIS
into a mobile system that allows a user to enjoy immersive VR
exerciseexperienceanywhere.

VirtualRealityinSports SWOTAnalysis
ThePotentialUsefulnessofVirtualRealitySystems
forAthletes:AShortSWOTAnalysis
Peter Düking, Hans-Christer Holmberg and Billy Sperlich
Integrative & Experimental Exercise Science & Training, Institute for Sport Sciences, University of Würzburg, Würzburg, Germany; Swedish Winter Sports Research Centre, Mid Sweden University, Östersund,
Sweden; School of Sport Sciences, UiT The Arctic University of Norway, Tromsø, Norway; Schoolof Kinesiology, University of British Columbia, Vancouver, BC, Canada
Front. Physiol., 05March 2018 https://doi.org/10.3389/fphys.2018.00128
Virtual reality (VR) systems (Neumannetal.,2017), which are
currently receiving considerable attention from athletes,
create a two- or three-dimensional environment in the form
of emulated pictures and/or video-recordings where in
addition to being mentally present, the athlete even often
feels like he/she is there physically as well. As she/he
interacts with and/or reacts to this environment, movement
is captured by sensors, allowing the system to provide
feedback.
As with every newly evolving technology related to human
movement and behavior, it is important to be aware of the
strengths, weaknesses, opportunities and threats (SWOT)
associated with the use of this particular type of technology.
SWOT analyses are widely utilized for strategic planning of
developmental processes (PicktonandWright,1998;
TaoandShi,2016) and it is of great interest to consider
whether VR systems should be adopted by athletes or not.
Aspects more inherent to the employed technologies of VR
systems, and aspects more related to the application of VR
systems with athletes are considered as
strength/weaknesses and opportunities/threats,
respectively. Analogously, SWOT analysis concerning
another emerging technology involving sensors of individual
parameters (i.e., “implantables”) has been performed (
Sperlichetal.,2017).

VirtualReality engagement ideas NaturalisticSetting
ZenART VR Experiences
https://www.zenartvr.com/
Photorealisticrenderings
for the virtual reality?
GeorgeMaestriatAutodeskUniversity
https://www.autodesk.com/autodesk-university/class
/Approaching-Photorealism-Virtual-Reality-2018
ImmersiveRehab
Finalistcategory: DigitalHealthAward
https://www.tech4goodawards.com/finalist/immersive-rehab/
vs
Brackeys
PublishedonJan25,2017
https://youtu.be/IlKaB1etrik

Virtual Reality Graded ExposureTherapy forChronicLow BackPain: A
PilotStudy withHTC Vive /Unity
https://projekter.aau.dk/projekter/files/281189247/MTA181038_Virtual_Reality_Graded_Exposure_Therapy_for_Chronic_Low_Back_Pain_A_Pilot_Study.pdf
With the advent of affordable high performance virtual reality system, we
investigate the feasibility and acceptability of a *virtual reality game for
treatment ofchroniclow backpain*.
Methods: We used graded activity, biofeedback, and gamification
principles to create a virtual reality dodgeball game where patients have to
pick up balls and hit enemies. We create a full body tracking solutionsuch
that we can tailor the game to the individual patients range of motion. The
game is further created with feedback froman expertinpainrehabilitation.
Results: The game is tested with experts, patients, and a healthy sample.
The experts were interviewed on feasibility and usability, patients on
acceptability, and healthy participants on general usability. The findings
showed that the game in a clinic setting was very feasible, and patients
were high encouraged by the game,and moving more thanbaseline.
Conclusion: We found that the game could be used in a clinic setting, and
patients are very willing to play as well as finding it fun, while not increasing
or decreasing back pain, and provides suggestions for future
improvements._

AddingFeedbacktoVR finger/headtracking
SaeboGlove orthosis with sensors to track grasp interactions
https://clinicaltrials.gov/ct2/show/NCT03036033
https://www.uhmb.nhs.uk/media-centre/latest-news/86k-revoluntionary-equipme
nt-will-benefit-stroke-patients/
A commercial SaeboGlove orthosis was fitted with wrist and finger motion
sensors to permit tracking of finger joint angles during grasp-release interactions
with a virtual environment. The sensors were attached to the existing tensioner
band hooks on the dorsal side of each glove finger. An electronics enclosure
mounted to the palmar side of the SaeboGlove’s plastic wrist splint processes the
sensor data and transmits information to a personal computer (PC) that hosts the
modified SaeboVR software. Data from both the SaeboGlove-integrated
sensors and from a Kinect sensor were used by a custom motion capture
algorithm, which employs a human UE kinematics model to produce real-time
estimatesofarm, wrist, and finger joint angles.
UpperExtremityFunctionAssessmentUsingaGloveOrthosisandVirtual
RealitySystem RichardJ.Adams, AllisonL.Ellington, KateArmstead March2,2019
https://doi.org/10.1177/1539449219829862
TheChangingLandscapeof OccupationalTherapyInterventionand
ResearchinanAgeof UbiquitousTechnologiesLiliLiu, AlexMihailidis
March19,2019 Editorialhttps://doi.org/10.1177/1539449219835370
When voice-controlled speakers such as Amazon Alexa and Google Home are marketed to
the general population, one may ask if they are also immediately useful to persons living with
disabilities, and as such, can they be considered as assistive devices? Furthermore, we will
quickly emerge as a generation where there may be a robot in everyone’s home. These
assistive and social robots will provide assistance across a variety of activities, from keeping a
home clean to supporting older adults through common activities of daily living. More
importantly, the cost of these robots is significantly being reduced each year, which is making
themmoreviableasan assistivetechnology

Finger/headtracking Do youneedextrasensors anymore?
OculusQuest'shandtrackingisa new
levelofVRimmersion 27 Sept 2019
It couldbe huge formobile virtual reality.
https://www.engadget.com/2019/09/27/oculus-quest-hand-tr
acking-hands-on
https://youtu.be/2VkO-Kc3vks
Today, we’remarkinganother importantmilestone with the
announcement of handtrackingonOculusQuest, enabling natural
interaction in VR using your own hands on an all-in-one device — no
extrahardwarerequired.
This is an important step, not just for VR, but for AR as well. Hand
trackingon Questwillbereleasedasanexperimentalfeaturefor Quest
ownersandadeveloperSDKinearly2020. Facebook CEO Mark Zuckerberg used the company’s Oculus Connect
developer conference in San Jose Wednesday to announce 2 major
updatesforthecompany’s Oculus QuestVR
https://variety.com/2019/
digital/news/oculus-que
st-hand-tracking-rift-pc-l
ink-1203348827/

https://doi.org/10.1080/21520704.2018.1563573 UKcompanyusingVR forfootball player
rehab https://rezzil.com/
JonathanM.Bird
BrunelUniversity, London, UK
https://bura.brunel.ac.uk/handle/2438/18291
Readyexerciser one: examining theefficacyof
immersivetechnologiesintheexercise
domain
Stealideas fromsportspsychologyfor engagement
GoaltenderVR; FinalSoccerVR; LumenVR; RelaxVR; 3DOrganonVRAnatomy

Stealideas frombehaviorial changestudies
1 - Rehabilitation for domestic abusers In this study, Mel Slater and
his team allow convicted offenders to step in the body of a victim of
domestic abuse. Compared to a control group, participants in the VR
experience improved their ability to recognize fearful female faces. Early
evidence suggests a decrease in recidivism although it is to early to
conclude that there isan absolute correlation.
2 - VR & Implicit Racism Bias Implicit racial bias plays can play a
crucial and dangerous role in a legal system that relies on a jury's
judgment. In this study, Natalie Salmanowitz puts different groups of
participants in either Caucasian or Black bodies then asks them to
evaluate a mock crime scenario. Not only did the people who embodied
a black avatar produced significantly lower implicit racial bias but they
were also more conservative when evaluating guilt, rating vague
evidence as less indicative of guilt and rendering more 'Not Guilty'
verdicts.
3 -VR & Attitude towards Homelessness This study, ran by a team
at Stanford University, looks at measuring the long-term behavioral
impact of taking one's perspective in VR vs the traditional or desktop
computer-based methods. In this case, the perspective taken was one of
someone becoming homeless. The results show that a significantly
higher number of participants in the VR condition signed a
petition supporting affordable housing for the homeless, despite very
little differences between the groups when it came to self-reported
measures of empathy. This is a striking example of sustained behavioral
change on a subconsciouslevel.
Published on LinkedIn, September 25, 2019 - Christophe Mallet
Unlocking Human Potential in the workplace with BODYSWAPS® ¦ AR/VR/MR Entrepreneur ¦ Immersive Learning Specialist

SocialMediaAutomation
”VerifyingforInstagramaudiencethatyouactuallywenttothegym”
Wearables,SocialNetworkingandVeracity:The
BuildingBlocksofaVerifiedExerciseApplication
Chiung Ching Ho ; Mehdi Sharif MultimediaUniversity,Cyberjaya63100,Selangor,Malaysia
20144th International Conference on Artificial Intelligence with
Applicationsin Engineeringand Technology
https://doi.org/10.1109/ICAIET.2014.28
Research and development of exercise
recognition applications have predominantly
focused on motion related exercise, with not
much emphasis on weight lifting exercise.
At the same time, while such applications
supports the posting of completed exercise
session on social network, the veracity of the
post is entirely determined by the user of the
application. In this paper, we present the building
blocks for a weight lifting application. It recognizes
and counts the number of repetitions of a weight
lifting exercise, andsubsequently posts it on the
user's behalf, thus ensuring the veracity of
the post. Our empirical results demonstrate the
potentialof such anapplication.
Feelinggreatabout thewaywelook
andbrowsingInstagramarenot,
generally,twothings thatgohandin
hand.It’sno surprisethatastudy
releasedearlierthisyearby
theRoyalSociety ForPublicHealth
foundthatthesocialmediaapp is,in
fact,theworstofallwhen itcomesto
negativelyimpactingon young
people’smentalhealth.The
researcherscitedbodyimageasa
keyfactorin theirfindings, aswellas
anxiety,depression andloneliness.
GeorgieOkell
https://graziadaily.co.uk/life/real-life/gym-selfie-didnt-happen-instagram-ruining-exercise/

Human Factors
“Allabout”
keepingthe
motivationhigh
Adherence, engagement,
progress visualization,
gamification, etc.

Visualizeprogress thelow-hangingfruitforsomemotivation
Do you want to record this just for the fun
of recording, or is the recording used in
motivating way?
See for comparison, e.g.
Us' em: The user-centered design of a device
for motivating stroke patients to use their
impairedarm-handindailylifeactivities
PMarkopoulos, AAA Timmermans
https://doi.org/10.1109/IEMBS.2011.6091283
Citedby24 -Relatedarticles
“Therapists and patients were asked to rate the
products using the CEQ inventory [Devilly and Borkovec2000]
for measuring credibility and expectations from the
device as an instrument for therapy; the scores on this
scalecanrangefrom9to27.“
Gym Tonic-Exercise as Medicine
https://www.gymtonic.sg/pilot/gymtonic.html
PulseSync Pte Ltd, AB Hur Oy, Raisoft OyLtd, Lien
Foundation, KokkolaUniversityConsortium Chydenius/
University of Jyväskylä

ProgressVisualizationcompeteagainstyourselforyourpeers?
Rendering, by ML, an “extracted”
skeleton image as an overlay
over an actual 3D moving image
of a stroke patient in real-time
(checking for anomalous gait
kinematics).
https://react-fitness.com/interactive-fitness-eq
uipment/

Takingrehabilitationtopatients’homes
Home-basedRehabilitationWithANovel
DigitalBiofeedbackSystemversus
ConventionalIn-personRehabilitationafter
TotalKneeReplacement:afeasibilitystudy
Scientific Reportsvolume 8, Article number:11299(2018)
https://doi.org/10.1038/s41598-018-29668-0
“This is the first study to demonstrate that
a digital rehabilitation solution can
achieve better outcomes than
conventional in-person rehabilitation,
while less demanding in terms of human
resources.
We have tested a novel digital
biofeedback system for home-based
physical rehabilitation (SWORD). Using
inertial motion trackers, this system
digitizes patient motion and provides real-
time feedback on performance through a
mobile app. It also includes a web-
based platform that allows the clinical
team to prescribe, monitor and adapt the
rehabilitationprocessremotely.
(A)MotionTrackerSetup.(B-C)MobileApp.(D-E)WebPortal
“IWasReallyPleasantlySurprised”:
FirsthandExperienceandShiftsinPhysical
TherapistPerceptionsofTelephone‐
DeliveredExerciseTherapyforKnee
Osteoarthritis–AQualitativeStudy BelindaJ.
Lawford ClareDelany Kim L.Bennell RanaS.Hinman
08 June 2018 https://doi.org/10.1002/acr.23618
Implementationofperson centredpractice‐
principlesandbehaviourchange
techniquesaftera2 daytrainingworkshop:‐
Anestedcasestudyinvolving
physiotherapistsBelindaJ. Lawford KimL.Bennell
JessicaKasza Penny K.Campbell JanetteGale
CarolineBills RanaS.Hinman 12April 2019
https://doi.org/10.1002/msc.1395
Medium-Term Outcomesof DigitalVersus
ConventionalHome-Based RehabilitationAfter
TotalKneeArthroplasty:Prospective,Parallel-
GroupFeasibilityStudy
FernandoDiasCorreia, MD SWORD Health
http://dx.doi.org/10.2196/13111 |

Thelessrequiredsensorstheeasiertodeploythesystemathome
Note! Some “extra” hardware might be still required for clinically useful system to-be-built
DesignandAnalysisof CloudUpperLimb Rehabilitation
SystemBasedonMotionTrackingfor Post-Stroke
Patients JingBai,AiguoSong,HuijunLi Appl.Sci.2019,9(8),1620
https://doi.org/10.3390/app9081620-Citedby1
In order to improve the convenience and practicability of home
rehabilitation training for post-stroke patients, this paper presents a
cloud-based upper limb rehabilitation system based on motion tracking. A 3-
dimensional reachable workspace virtual game (3D-RWVG) was
developed to achieve meaningful home rehabilitation training. Five movements
were selected as the criteria for rehabilitation assessment. Analysis was
undertakenoftheupper limbperformanceparameters
Target-Specific ActionClassificationforAutomated
Assessment of HumanMotorBehaviorfromVideo
BehnazRezaei,YiorgosChristakis,BryanHo,KevinThomas,KelleyErb,
SarahOstadabbasandShyamalPatelAugmentedCognitionLab (ACLab),NortheasternUniversity;DigitalMedicine&
TranslationalImaginggroup,Pfizer;Neurology Department,TuftsUniversitySchoolofMedicine; Department ofAnatomy & Neurobiology,BostonUniversity SchoolofMedicine
(20Sep2019)https://arxiv.org/abs/1909.09566
In this paper, we present a hierarchical vision-based behavior
phenotyping method for classification of basic human actions in video
recordings performed using a single RGB camera. Our method addresses
challenges associated with tracking multiple human actors and
classification of actions in videos recorded in changing environments with
differentfieldsofview.
The work presentedhereinfocusedonthe classification of basicpostures
(sitting, standing and walking) and transitions (sitting-to-standing and
standing-to-sitting), which commonly occur during the performance of
many daily activities and are relevant to understanding the impact of
diseases like Parkinson’s disease and stroke on the functional ability
ofpatients.
This has laid the foundation for future research efforts that will be directed
towards detecting and quantifying clinically meaningful information
like detection of emergency events (e.g. falls, seizures) and assessment of
symptom severity (e.g. gait impairments, tremor) in patients with
various mobility limiting conditions. Lastly, the code and models developed
during this work are being made available for the benefit of the broader
researchcommunity.

HowtoSelectBalanceMeasures Sensitive
toParkinson’sDiseasefromBody-Worn
InertialSensors—SeparatingtheTrees
from theForest
Sensors2019,19(15),3320;
https://doi.org/10.3390/s19153320
This study aimed to determine the
most sensitive objective measures of
balance dysfunction that differ
between people with Parkinson’s
Disease(PD) and healthy controls.
"Measures from the most sensitive
domains, anticipatory postural
adjustments (APAs), and Gait, were
significantly correlated with the
severity of disease and with patient-
related outcomes. This method
greatly reduced the objective
measures of balance to the most
sensitive for PD, while still capturing
four of the fivedomains of balance."
Youstill need theresearch forthe bestmetricsthatyou wanttotrackwithdeep
learning nomagicbulletofgettingclinicallyrelevant predictionsfromcrappydata→ I adopt the same here for

AI ModelCanRecommendtheOptimalWorkout April 24, 2019
https://news.developer.nvidia.com/ai-model-can-recommend-the-optimal-workout/
To help deliver more personalized workout recommendations,
University of California, San Diego researchers Jianmo Ni, Larry
Muhlstein and Julian McAuley developed a deeplearning-based
system to better estimate a runner’s heart rate during a
workoutand predicta recommended route.Theworkhasthe
potential to help fitness tracking companies and mobile app
developersenhancetheirappsanddevices.
Once trained, the algorithm relies on the GPU to generate the
recommended route. The system is able to detect hills and
obstacles that might alter a user’s heart rate. The tool can also
recommend alternate routes for users who are working
towardsaspecificheartrate.
Example Probably goodforcasualrunnerstohave “automatic
alternate”routesforsomevariations,but beyond?
Model structure for workout profile forecasting (FitRec) and short term prediction (FitRec-Attn). FitRec
contains a 2-layer stacked LSTM and FitRec-Attn has an encoder-decoder module with dual-stage attention.
Thefinaloutputsarecolored inblue. https://cseweb.ucsd.edu/~jmcauley/pdfs/www19.pdf

RecommendationEngine for
‘PrecisionRehabailitation’
Summary
Beginnerathletes
No way really of knowing if the
recommendations make sense
without a human therapist Needs→ I adopt the same here for
good clinical validation studies before
can be taken byskepticaltherapists
Advanced Athletes
The End-user will want to return your
crappy device if it makes stupid
recommendations Your business/→ I adopt the same here for
service won’tsucceed
Mightbesufficientjusttoquantify
ifthemovement is“textbook-like”
forexercise naïve subjects
Youwanttoquantifymuscle
activation (i.e.muscle-mind
activation),and trackthisover
timealongrecoveryparameters

With theproper pathology-specific exercises found thinkabouthowtovisualizethe
progressforthepatients
Homeself-training:Visualfeedbackfor
assistingphysicalactivityforstrokesurvivors
RenatoBaptistaetal.(2019) University of Luxembourg
https://doi.org/10.1016/j.cmpb.2019.04.019
A novel low-cost home-based training system is
introduced. This system is designed as a composition of
two linked applications: one for the therapist and another
one for the patient. The therapist prescribes personalized
exercises remotely, monitors the home-based training
and re-adapts the exercises if required. On the other side,
the patient loads the prescribed exercises, trains the
prescribed exercise while being guided by color-based
visual feedback and gets updates about the exercise
performance. To achieve that, our system provides three
main functionalities, namely: 1) Feedback proposals
guiding a personalized exercise session, 2) Posture
monitoring optimizing the effectiveness of the session, 3)
Assessmentofthequalityofthemotion.
●
Anovellow-costhome-basedtrainingsystem
dedicatedtostrokesurvivorsisintroduced.
●
Our systemiscomposedoftwolinkedapplications:
therapistandpatientapplications.
●
Theprescriptioniscreatedandpersonalizedinthe
therapistapplication.
●
A color-based visual feedback tool is proposed to
guidethepatientswhiletraining.

Howtoquantifyadherenceandengagement?
VerificationofaPortableMotionTrackingSystemforRemote
Managementof PhysicalRehabilitationoftheKnee Sensors2019, 19(5),
1021;https://doi.org/10.3390/s19051021
(ThisarticlebelongstotheSpecialIssue GyroscopesandAccelerometers)
“We developed a remote rehabilitation management
system combining two wireless inertial measurement units
(IMUs) with an interactive mobile application and a web-based
clinician portal (interACTION). However, in order to translate
interACTION into the clinical setting, it was first necessary to
verify the efficacy of measuring knee motion during rehabilitation
exercises for physical therapy and determine if visual
feedbacksignificantly improvesthe participant’s ability
toperformthe exercisescorrectly.
Exercises were recorded simultaneously by the IMU motion
tracking sensors and a video-based motion tracking
system (OptiTrack, running the Motive: Tracker software was
utilized as the “gold standard [Thewlis et al. 2013, Carse et al.2014]
). Validation
showed moderate to good agreement between the two systems
for all exercisesandaccuracywaswithinthreedegrees.Basedon
custom usability survey results, interACTION was well
received. Overall, this study demonstrated the potential of
interACTION to measure range of motion during rehabilitation
exercises for physical therapy and visual feedback
significantly improved the participant’s ability to
performtheexercisescorrectly.
(A) Yost Lab’s two 3-Space Bluetooth sensors is a 3D printed case designed to align the sensors during
alignment, (B) Padded elastic straps secured on the thigh and shank, Cary, (C) Screenshot of the mobile
application screen that providesthe participant with visual feedback.

AdherencedependsalotontheengagementandrehabsystemUX
Adherencemonitoringofrehabilitation
exercisewithinertialsensors:Aclinical
validationstudysLuckshmanBavana, Karl Surmacz,
David Beard, Stephen Mellon, Jonathan Rees(Nuffield
Department of Orthopaedics, Oxford) Gait& PostureVolume
70,May2019, Pages 211-217
https://doi.org/10.1016/j.gaitpost.2019.03.008
“Aims to evaluate the feasibility of using a single
inertial sensor (MetaMotionR, MbientLab,) to
recognise and classify shoulder rehabilitation
activity using supervised machine learning
PatientInvolvementWithHome-Based ExercisePrograms:CanConnectedHealth
InterventionsInfluenceAdherence?sRob Argentet al., Beacon Hospital, UniversityCollege Dublin Beacon
Academ https://doi.org/10.2196/mhealth.8518
“Adherence to home exercise in rehabilitation is a significant problem, with estimates of
nonadherence as high as 50%, potentially having a detrimental effect on clinical outcomes. In this
viewpoint, we discuss the many reasons why patients may not adhere to a prescribed exercise
program and explore how connected health technologies have the ability to offer numerous interventions
to enhance adherence; however, it is hard to judge the efficacy of these interventions without a
robustmeasurementtool.”
“It is widely accepted that at present, there is no gold standard for the measurement of adherence to
unsupervised home-based exercise, as the significant proportion of outcome measures used in the
literature rely on patient self-report and are therefore susceptible to bias [Bollenetal.2014]. In a
systematic review of 61 different self-reported outcome measures for adherence to home-based
rehabilitation, only two measures scored positively for a single psychometric property of validation [
Bollenetal.2014]. Furthermore, the outcome of any research studies using paper diaries or retrospective
recall has been called into question as it is highly prone to recall and self-serving bias [
Stoneetal.2003]. Equally, these measures make no allowance for the quality of performance, as
highlightedintheabovementioneddefinition.”
“Sensing platforms such as the use of IMUs or motion capture camera are rapidly advancing and
couldbe an opportunitytomake amoreobjective assessmentofadherence,continuouslytracking motion
data obtained from an individual [Rizketal.2013; Oeschetal.2017]. However, the use of these devices to
measure adherence is questionable as they arguably influence/enhance adherence itself by means of
the user knowingthat they are beingrecorded. In thisway the end pointisinfluenced greatly by the
measurement strategy, leading to questionable results as the patient no longer has the choice on whether
to adhere [Bollenetal.2014].Regardless of the challengeswith accurately measuring adherence, itis clear
thatthereareproblemswithadherencetoprescribedexerciseinthehomesetting.”

Therapistinloopwithroboticrehabilitation
LearningandReproductionofTherapists
Semi-Periodic Motions duringRobotic
Rehabilitation
CarlosMartinez andMahdi Tavakoli
Robotica(21May2019)
https://doi.org/10.1017/S0263574719000651
The demandfor rehabilitation serviceshasincreased in
recent years due to population aging. Due to the
limitations of therapist’s time and healthcare resources,
robot-assisted rehabilitation is becoming an
appealing, powerful, and economical solution. In this
paper, we propose a solution that combines Learning
from Demonstration (LfD) and robotic
rehabilitation to save the therapist’s time and
reduce the therapy costs when the therapy
involvesperiodicorsemi-periodicmotions.
We begin by modeling the therapist’s behavior (a
periodic or semi-periodic motion) using a Fourier
Series (FS). Later, when the therapist is no longer
involved, thesystemreproducesthelearned behavior
modeled by the FS using a robot. A second goal is to
combine the above with Gaussian Mixture Model
(GMM) and Gaussian Mixture Regression (GMR) to
obtain a more flexible and generalizable reproduction
of the therapist’s behavior. This algorithm allows
learning and imitating repetitive movement tasks. Our
experimental results show the application of these
algorithmstorepetitivemotiontask.
Therapists have the knowledge and skill to determine the required assistance or resistance for a
given patient in a given phase of recovery and are also able to modify or adapt the given task based
on patients necessities. Because robots do not have this ability, a therapist has to be involved at
least for a short duration at the beginning of rehabilitation therapy. In this paper, we propose to use
LfD as a solution to reprogram rehabilitation robots based on observing a brief window of
therapist-patient interaction. The proposed LfD algorithm allows the robot to be reprogramed as a
therapist moves the robot while it is in a passive (compliant) mode; this teaching method is
known as kinesthetic teaching (Lee et al. 2012) Cited by 29
.

Introducingrobotic upper limb training into
routineclinical practice for stroke survivors:
Perceptionsof occupational therapistsand
physiotherapists (July 2019)
https://doi.org/10.1111/1440-1630.12594
"Therapists’ expressed their optimism
towards the introduction of RT-UL but
believed successful implementation
would be primarily dependent on the
availability of clinical leadership, training
anda suitable client mix.
Conclusion: Therapists perceived that
RT-UL would provide opportunity for
increased upper limb practice
particularly for patients with severe
upper limb impairment. To facilitate
implementation, support of RT-UL
should come from both management
and clinical leaders and training include
RT-UL efficacy, device functionality and
patient suitability. The availability of a
single RT-UL device in a workplace may
create unique interdisciplinary and
logistical challenges."
Robotassistedtrainingfortheupperlimb
afterstroke(RATULS):amulticentre
randomisedcontrolledtrial
Helen Rodgers et al. (Lancet 2019)
https://doi.org/10.1016/S0140-6736(19)31055-4
Loss of arm function is a common problem after stroke. Robot-
assisted training might improve arm function and activities of
daily living. We compared the clinical effectiveness of
robot-assisted training using the MIT-Manus robotic
gym with an enhanced upper limb therapy (EULT) programme
based on repetitive functional taskpractice and withusual care.
Robot-assisted training and EULT did not improve upper limb
function after stroke compared with usual care for patients with
moderate or severe upper limb functional limitation. These
results do not support the use of robot-assisted training
as provided in this trial inroutine clinical practice.
Therapistsperceiverobotictherapy well, but isit really effectice?

ParasiticBody: A virtual reality system to study the collectionofvisualfeedback from roboticarms Recent advancementsin robotics
have enabled the development ofsystemsto assist humansin a varietyof tasks. Atype ofrobotic system that hasgained substantial popularityover the
past few yearsiswearable roboticarmsremotelyoperatedbya thirdparty. https://techxplore.com/news/2019-09-parasitic-body-virtual-reality-visual.html
RyoTakizawaetal.ParasiticBody:ExploringPerspectiveDependencyinaSharedBodywithaThirdArm, 2019IEEEConferenceonVirtualRealityand3DUser
Interfaces(VR) (2019). DOI:10.1109/VR.2019.8798351
Could youoptimizetherobotictreatmenttobe actually usefulthen?

StrokeRehab and SportScience/Performingarts veryclose toeach other method-wise
”Sensorimotortraining”
AWearableSensor-BasedExercise
BiofeedbackSystem:MixedMethods
EvaluationofFormulifts
O'ReillyMA,SlevinP,WardT,CaulfieldB
https://doi.org/10.2196/mhealth.8115
Thispaper isin the followinge-collection/theme issue:
mHealth for Wellness, Behavior Change and Prevention | Mobile Health (mhealth)
Human Factors and Usability CaseStudies | Usabilityand userperceptions of mHealth
Design and Formative Evaluation of Mobile Apps | Wearable Devices and Sensors
Formulift is a newly developed mobile health (mHealth)
app that connects to a single inertial measurement
unit (IMU) worn on the left thigh. The IMU captures
users’ movements as they exercise, and the app
analyzes the data to count repetitions in real time and
classifyusers’exercisetechnique.
The aim of this study was to assess the Formulift system
with three different and realistic types of potential users
(beginner gym-goers, experienced gym-goers, and
qualified strength and conditioning [S&C]
coaches)
This study demonstrated an overallpositive evaluation of
Formulift in the categories of usability, functionality,
perceived impact, and subjective quality. Users also
suggested a number of changes for future iterations of
the system. These findings are the first of their kind and
show great promise for wearable sensor-based
exercisebiofeedbacksystems.
Unravelingmysteriesofpersonal
performancestyle;biomechanicsof left-hand
positionchanges(shifting)inviolin
performance
PeterVisentin,ShimingLi,GuillaumeTardif,Gongbing
Shanhttps://peerj.com/articles/1299/
Instrumental music performance ranks among the
most complex of learned human behaviors. It requires
intricate motor skills, perception and adaptation in a
temporal endeavor, and sensory and neural discrimination
thatchallengesthelimitsofhuman cognition
Given successesthat have been achieved by applying
scientific methods in athletic training, it seems
logical to adapt these to the context of music
performance. In a 2002 comprehensive review,
Kennell acknowledged “growing professional interest
in applying the tools of systematic research to the
context of studio instruction in music education
research” (Kennell,2002). None of the studies cited
addressed any aspect of teaching the
biomechanical skills requisite for successful
musicalperformance(Flohr &Hodges,2002).
A 3-D motion-capture system was used to measure
full-body movement using 68 reflective markers—39 on the
body, 22 on the left hand, 4 on the violin and 3 on the bow. A
twelve-camera VICON MX40 motion capture system
(VICON Motion Systems, Oxford Metrics Ltd., Oxford,
England)trackedthemarkersatarateof200frames/s.
The study used methods from movement science to
examine timing elements and motor control strategies
during shifting, a skill vital in violin performance. It
contributes tofundamentalunderstanding ofthe skilland
discusses elements of individualization among
subjects in terms of anthropometry and the
strategic use of motor behaviors developed through
lengthy practice. Finally, it considers the implications of
these in terms of the aural result. In doing so, the
current study points in the direction of a research inquiry
model that might meaningfully influence music
pedagogy and provides a basis for future studies that
examine the manipulation of motor behaviors as a
foundationalelementofartistryinmusicperformance.

Hardware
(sensors and
interfaces)
forclinical
stroke
rehabilitation

Front.Physiol.,28June2018| https://doi.org/10.3389/fphys.2018.00743
ACriticalReviewofConsumerWearables,MobileApplications,andEquipmentfor
ProvidingBiofeedback,MonitoringStress,andSleepinPhysicallyActivePopulations
JonathanM.Peake, GrahamKerr and JohnP.Sullivan
Brisbane,QLD,Australia
https://doi.ieeecomputersociety.org/10.1109/TMSCS.2017.2675888
Significanceof NanomaterialsinWearables:AReview on
WearableActuatorsand Sensors(2018)
https://doi.org/10.1002/adma.201805921

Optical
Motion
Capture
i.e.
Computer
Vision

Throwinmoretoysforgaitanalysis
Kinematic analysis (Motioncapture)andinertialmovementunits(IMUs)
formorefine-levelquantificationofmovement
Monitoringgaitkinematicsduringtherapyofacutespinalcordinjury
(SCI) andstrokepatientsandformulatebetterpredictorsofrecovery
http://faculty.engr.utexas.edu/rewire/rewire/book/longitudinal-gait-analy
sis-using-imu-sensors
Feasibility study of
using aMicrosoft
Kinect forvirtual
coaching of
wheelchair
transfer
techniques “Gold
Standard” with Vicon
motion capture
systems
https://doi.org/10.1
515/bmt-2015-02
06
Gait Analysis& Rehabilitation
ViconprovidesaClinicallyValidatedsolutiondesignedspecificallytosuityour
needsinanygaitanalysisorrehabilitationenvironment.
Posture,Balance andMotor Control
Viconsystemscanbeusedtomeasureor givereal-timefeedbackonthe
movementsofthewholebodyor asinglepart,includingdetailedhands,face,
feetandspineacrossdifferentapplications.For example,strokerehabilitation,
postureanalysis,balancestudiesandreachingstudies.
https://www.vicon.com/motion-capture/life-sciences
https://www.vicon.com/press/2018-02-20/vicon-integrates-inertial-tracking-i
nto-the-optical-world

Otheralternatives for expensivemotioncapture
Affordable gaitanalysisusing augmented
reality markersGergelyNagymáté,RitaM.Kiss
February14,2019
https://doi.org/10.1371/journal.pone.0212319
Citedby1 -Relatedarticles
Calibrationofanatomicalpointsusingthe
calibrationpointer.
There are initiatives where open source solutions are provided to replicate the stereophotogrammetry
based functionality of motion capture systems with consumer grade cameras. Jackson et al. [10]
offers a complex solution for necessary camera calibration and the synchronization of video inputs from
multiple cameras. This approach is based on stereophotogrammetry, where the identifiable points of the
tracked object have to be seen from different angles by multiple cameras. Another image processing
approach is homography, which relates the transformation between two planes [11]. This is used in
photographyforpanoramapicturestitchingorperspective correctionandisalsousedin augmentedreality
(AR) to estimate camera pose from coplanar points and vice versa. It can identify rotations and
translations (3D kinematics) of an AR marker relative to the camera focus point and the image plane by how
the corners of the known geometry marker appear on the recorded image. Compared to continuously drifted
or zero corrected IMU-s, the 6 degree of freedom tracking of AR markers make them possible to track
the absolute position of external objects [12] and body segments if attached to them. Compared to
stereophotogrammetry basedalternatives [10], AR marker basedtrackingcanworkwith onecamera,
althoughin thiscasethemovementdirection can belimited(e.g.treadmillwalking).
AR was mostly mentioned so far in motion studies as a part of therapies [13], but not for the purpose of
biomechanical motion tracking. Ortega-Palacios et al. describe a gait analysis system with augmented reality,
but the localization of infra-red LED (light emitting diode) markers is still processed by
stereophotogrammetry [14]. Sementille et al. used actual augmented reality markers to track the position of
jointson avery simplifiedanatomicalmodel[15].Noneoftheaboveresearchworksvalidatedthedataacquired
usingaconventionalmotion analysissystem.
The first aim of this research is to present a novel approach for gait analysis with a single commercial
action camera using augmented reality markers based on the approach of tracking body segments by
marker rigid bodies [3]. Therefore, no simplification of the anatomical model is required, a full six degree of
freedom kinematic analysis of each body segment and joint is possible using conventional or open-source
motion analysis solutions such as OpenSim (NIH Center for Biomedical Computation, Stanford University,
http://opensim.stanford.edu/).
The second aim of the paper is to validate a possible implementation of the proposed approach by
simultaneous measurements with a conventional motion capture system on treadmill gait trials of healthy
subjects of varying age at different walking speeds, followed by comparing the coordinates of the tracked
virtualanatomical pointsandcalculationsforcomparing angularand spatialgait parameters.

SmartphoneRGB(D) asthemostaccessibleof course
ValidityandReliabilityof StandingPosture
MeasurementsUsingaMobileApplication
BreannaBerryHopkinsetal. (2019)
JournalofManipulativeandPhysiologicalTherapeutics
https://doi.org/10.1016/j.jmpt.2019.02.003
The purpose ofthis study wasto evaluate the validity and
reliability of standing posture assessments in
asymptomatic men using the PostureScreenMobile
(PSM)iOSapplication.
SquatScreen is a professional HIPAA compliant application geared for Strength and
Conditioning coaches, Personal Trainers, Chiropractors, Physical Massage Therapists, and
other fitness professionals who wish to quickly and objectively evaluate the functional
movementforclients.https://itunes.apple.com/gb/app/squatscreen/id1249748805
The following 10 measurements using the PSM
app were compared to the criterion VICON 3-
dimensional analysis: from the frontal plane,
shift and tilt of the head, shoulders, and hips; and
from the sagittal plane, shift of the head, shoulders,
hips, and knees. We used Bayesian methods to
analyze the data.
The use of the PSMappintroducedsignificant bias in postural measurements in
the frontal and sagittal plane. Until further research reports additional validity
and reliability data of the PSM app, we suggest caution in the use of PSM
appwhenhighlyaccurate posturalassessments arenecessary.

Quantifying Squatformforinjuryprevention withcamera
TemporalDistanceMatricesforSquat
Classification
RyojiOgata,Edgar Simo-Serra,SatoshiIizuka,HiroshiIshikawa;The
IEEE ConferenceonComputer VisionandPatternRecognition
(CVPR)Workshops,2019,pp.0-01
http://openaccess.thecvf.com/content_CVPRW_2019/html/CVSpo
rts/Ogata_Temporal_Distance_Matrices_for_Squat_Classification_
CVPRW_2019_paper.html
When working out, it is necessary to perform the same action
many times for it to have effect. If the action, such as squats or
bench pressing, is performed with poor form, it can lead to
seriousinjuriesin thelongterm.
With the prevention of such harm in mind, we present an action
dataset of videos where different types of poor form are
annotated for a diversity of subjects and backgrounds, and
propose a model for the form-classification task based on
temporaldistancematrices,both inthecaseof squats.
We first run a 3D pose detector, then normalize the pose and
compute the distance matrix, in which each element
represents the normalized distance between two joints. This
representation is invariant under global translation and rotation,
as well as robust to individual differences, allowing for better
generalization to real world data. Our classification model
consists of a CNN with 1D convolutions. Results show that our
method significantly outperforms existing approaches for the
task.
Failure cases. Warped Backis detected
even though thebackisin fact round. Thisis
mad difficult because there isnokeypointin
the middle of the back

MultiqualityOptical Motion capture Simultaneous measurement with all the devices
”Deeply-supervisednets” approach CYLee et al. 2015

Multimodal / “multiquality”model
“Optical-only” approach may leavesomeproblems resolve ambiguities with other modalities such as IMU/ IMUsuits
1
2
3
4
5
Multiquality
Optical
Motion
capture
v
Deep Full-Body Motion Network fora SoftWearableMotionSensing Suit
https://doi.org/10.1109/TMECH.2018.2874647
1
2
SingleInertialMeasurementUnit(IMU)
+ faster to setup and easier to use, with lower cost
- not as accurateas multisensor suit
http://doi.org/10.1136/bmjopen-2018-026326
‘GoldStandard’(IMU)
Mightresolvesome ambiguitiesfromoptical
motiontracking, whileoverall accuracy islowerthan
“optical groundtruth”?
+

Multimodal / “multiquality”model Thinkalsoabout “auxiliarymeasures” that allow youtoget
betterqualityrecordingswhichyou wouldnot intuitivelyassociatewithmotionquantifation.
I.etrytoquantifyartifacts and confoundingfactors aswell
1
2
3
4
5
v
1
2
+
Occlusions
Morecameras?
Deep learning?
Shinysurfaces
Polarization measurement?
Background/
Foregroundseparation
(“image matting”)
Optimize sensor
and illumination placement?
Moresuitableforindustrialrobotics
applicationsthogh
SoftTissueArtifacts
Algorithmiccompensation
More rigid suits?
Innovations inthe materials?

Oranalternative wayto see it is tohavethe “garbage
in” reduced withthe high-end device
supervision fromthemodelingpipeline
Inductiv developed technology
that uses artificial intelligence
to automate the task of
identifying and correcting
errorsindata*. Havingcleandata
is important for machine learning, a
popular and powerful type of AI that
helps software improve with less
human intervention.
* i.e. in order to train the “AI” to detect the errors, it is
useful to have some ground truth data, even if your
modelwasunsupervised
https://www.bloomberg.com/news/articles/2020-05-27/apple-
buys-machine-learning-startup-to-improve-data-used-in-siri?sr
nd=markets-vp&sref=0TyqkWgK

MotionModel “Inverteduse cases”
GlassesfortheThirdEye:Improvingthe
QualityofClinicalDataAnalysiswith
MotionSensor-basedDataFiltering
Jaeyeon Park, Woojin Nam, Jaewon Choi, Taeyeong Kim, Dukyong Yoon, Sukhoon Lee,
Jeongyeup Paek,JeongGil Ko AjouUniverisity,KunsanNationalUniversity,Chung-AngUniversity
https://doi.org/10.1145/3131672.3131690
Detect when patients move so that
their recordings are artifacted →
automatic signal quality
assessment (having some
uncertainty estimate for Bayesian
models)
BedsideComputerVision—Moving
ArtificialIntelligencefromDriver
AssistancetoPatientSafety
SerenaYeung, Lance Downing, Li Fei-Fei, Arnold Milsteino StanfordUniversity
https://doi.org/10.1145/3131672.3131690
+https://arxiv.org/abs/1708.00163
AI-based system using depth sensing
(for privacy concerns) for detecting
deviations from such essential
behavior as maintaining hand
hygiene.
Action recognition
useful beyond
physiotherapy as
well

Multimodal / “multiquality”model FinalOutput
Laboratory motionandforceplatedatacaptureoverlay.
“Predicting Athlete Ground ReactionForces and Moments fromSpatio-temporal Driven CNN Models,”
by William Johnson et al.
Magical
Model
Wehavea“fullbiomechanical
understanding”oftheindividual
patient/athlete
Nowyou“only”havetofigurehow
tousethisinformation,andhowto
studydesigns.Youmightwantto
●
Diagnose
●
Prognose
●
Designinterventionstogetthe
movementstosomedesired
target,i.e.howrehabfromstroke
optimally

Multimodal / “multiquality”model Finalmodel meetsreality
Magical
Model
Modeltraining
requires many
sensors tobebe
wornby many
subjects
Howmany
usersalready
haveFitbit
withexisting
data
collection
ecosystem?
Howmany people
couldbeplaying
someWiigame? Or
othervery
accessible
“quantification
method”
Toward personalized cognitive
diagnosticsofat-genetic-risk
Alzheimer’sdisease
Gillian Coughlan, AntoineCoutrot, Mizanur Khondoker, Anne-Marie
Minihane, HugoSpiers, and Michael Hornberger
PNAS publishedApril23,2019
https://doi.org/10.1073/pnas.1901600116

Whatdeviceswe coulduse?
IMUs InertialMeasurementUnits

IMUSsinexpensive|Thetechofthe“Fitbits”*ineverysmartphone
Low-end motion capture
systems, such as OptiTrack
(NaturalPoint, OR, USA), may
cost ~$15,000 USD; while high-
end video systems such as the
Vicon system (Vicon, Oxford, UK)
may run more than $200,000
USD [Thewlisetal.2013].
Recently, wearable inertial
sensors or inertial measurement
units (IMUs) have gained
attention in motion analysis for
their small size, low cost (usually <
$500 USD), and capability to
reveal 3D motion. IMUs typically
contain accelerometers,
gyroscopes, and magnetometers
conventionally used in navigation
systems. IMUs are becoming
well-established technology
for human gait studies [
Picerno2017].
FitbitAlta,SamsungGearFitSM-R350,Vidonn X6,Vidonn X6validated
withNaturalPointOptiTrackPrime13
http://doi.org/10.3390/proceedings2060197
*Somestepcountersmighthavejustxyz-accelerometersandnot“fullIMUs”
Adafruit 9-DOFAbsolute
Orientation IMUFusion
Breakout -BNO055
BoschSensortec
Best ofall you can get started in 10 minutes wit
hourhandytutorial onassembly, wiring, Circuit
Python& Arduino libraries, andProcessing gra
phical interface, and more!
Datasheet,EagleCADPCB
files,andFritzingavailablein
theproducttutorial
$34.95
https://www.mouser.fi/ProductDe
tail/Bosch-Sensortec/BNO055

IMUSsinrehabilitationcontext#1
MEMSInertialSensorsBasedGaitAnalysisforRehabilitation
AssessmentviaMulti-SensorFusion
SenQiu,LongLiu,HongyuZhao,Zhelong WangandYongmeiJiang
Micromachines2018,9(9),442;https://doi.org/10.3390/mi9090442
In this study, fluctuations of joint angle and asymmetry of foot elevation in
human walking stride records are analyzed to assess gait in healthy adults
andpatientsaffected withgait disorders.Thispaper aimstobuildalow-
cost, intelligent and lightweight wearable gait analysis platform based on the
emerging body sensor networks, which can be used for rehabilitation
assessment of patients with gait impairments. A calibration method
for accelerometer and magnetometer was proposed to deal with ubiquitous
orthoronalerrorandmagneticdisturbance.
Kneerangeof
motion(ROM)
recoveryhistory
beforeandafter
medicaltreatmentsfor
anarthropathypatient
andastrokepatient,
respectively.
UsingBody-WornSensorsforPreliminaryRehabilitation
AssessmentinStrokeVictimsWithGaitImpairment
SenQiu ;ZhelongWang; HongyuZhao;Long Liu;YongmeiJiang
UniversityofTechnology,Dalian,China
https://doi.org/10.1109/ACCESS.2018.2816816(2018)
This paper proposed a low-cost, intelligent, and lightweight wearable
platform for rehabilitation assessment in stroke victims with gait
impairment. The paper starts from the sensor physical properties and human
physiology structure, and aims to solve sensor drift problem by zero velocity
update algorithm. A complementary filter based on proportional integral
controller wasadoptedtoeliminatecomputationalerrors.
The concept of gait analysis (a) traditional observational gait analysis method
(b)twotypicalabnormalarch:strephenopodiaandstrephexopodia.
BodySensorNetworkbasedRobustGaitAnalysis:TowardClinical
andatHomeUsehttps://doi.org/10.1109/JSEN.2018.2860938 (2019)

IMUSsinrehabilitationcontext#2
UsingBody-WornSensorsforPreliminaryRehabilitation
AssessmentinStrokeVictimsWithGaitImpairment
SenQiu ;ZhelongWang; HongyuZhao;Long Liu;YongmeiJiang
UniversityofTechnology,Dalian,China
https://doi.org/10.1109/ACCESS.2018.2816816(2018)
Improving health is an important driving factor of sensor technology
applications. To meet the demands of precision medicine for medical
rehabilitation and elderly guardianship, using wearable sensors to get
kinematics, kinetics, and biochemical information has become an
interdisciplinary research hotspot recently. This paper proposed a low-cost,
intelligent, and lightweight wearable platform for rehabilitation assessment in
strokevictimswithgaitimpairment.
HipandtrunkkinematicsestimationingaitthroughKalmanfilter
usingIMUdataattheankle ABaghdadi,LACavuoto,JLCrassidis
IEEESensorsJournal,2018 https://doi.org/10.1109/JSEN.2018.2817228
The purpose of this paper is to provide a new method of estimating the hip
acceleration and trunk posture in the sagittal plane during a walking task
using an extended Kalman filter (EKF) and an unscented Kalman filter (UKF).
A comparison between these two estimation techniques is also provided.
Considering the periodic nature of gait, a modified biomechanical model
with Fourier series approximations are utilized as a priori knowledge. Inertial
measurement units (IMUs) are placed on the right side of the ankle, hip, and
middle of the trunk of twenty recruited participants, as input, a posteriori data,
andthegroundtruthforthemodel,separately.

IMUSsinforsportshealthexamination,andinjuryprognosis #1
Thevalueoftibialmountedinertialmeasurementunitstoquantify
runningkineticsinelitefootball(soccer)players.Areliabilityand
agreementstudyusingaresearchorientatedandaclinically
orientatedsystem
Tom Hughes, Richard K.Jones, ChelseaStarbuck,
Jamie C.Sergeant, Michael J. Callaghan
Manchester United Football Club,AON Training Complex / Universityof Manchester
JournalofElectromyographyandKinesiologyVolume44, February2019
https://doi.org/10.1016/j.jelekin.2019.01.001
In elite football, measurement of running kinetics with inertial measurement
units (IMUs) may be useful as a component of periodic health
examination (PHE). This study determined the reliability of, and agreement
between a research orientated IMU Delsys Trigno IM and clinically
orientated IMU system ViPerform for initial peak acceleration (IPA) and
IPAsymmetryindex(SI)measurementduringrunninginelitefootballers.
The use of IMUs to evaluate treadmill running kinetics cannot be
recommended in thispopulationasaPHEtesttoidentifyprognosticfactors
for injuriesorfor rehabilitationpurposes.
Reliability,ValidityandUtilityofInertialSensorSystemsforPostural
ControlAssessmentinSportScienceandMedicineApplications:A
SystematicReview
William Johnston, Martin O’Reilly, Rob Argent, BrianCaulfield
Insight Centre for Data Analytics, University College Dublin; Physiotherapy and Sports ScienceUniversityCollege Dublin; Beacon Hospital Dublin
SportsMedicine May2019
https://doi.org/10.1007/s40279-019-01095-9
This systematic review aims to synthesise and evaluate studies that have investigated
the ability of wearable inertial sensor systems to validly and reliably quantify
postural control performance in sports science and medicine applications. Future
research should evaluate the clinical utility of these systems in large high-quality
prospective cohort studies to establish the role they may play in injury risk
identification,diagnosisandmanagement.

Precision Physiotherapy & Sports Training: Part 1

Precision Physiotherapy & Sports Training: Part 1

Recommended

Recommended

More Related Content

What's hot

What's hot (16)

Similar to Precision Physiotherapy & Sports Training: Part 1

Similar to Precision Physiotherapy & Sports Training: Part 1 (20)

More from PetteriTeikariPhD

More from PetteriTeikariPhD (14)

Recently uploaded

Recently uploaded (20)

Precision Physiotherapy & Sports Training: Part 1