Mobile health is an emerging field which is attracting much
attention. Nevertheless, tools for the development of mobile health applications are lacking. This work presents mHealthDroid, an open source Android implementation of a mHealth Framework designed to facilitate the rapid and easy development of biomedical apps. The framework is devised to leverage the potential of mobile devices like smartphones or tablets, wearable sensors and portable biomedical devices. The framework provides functionalities for resource and communication abstraction, biomedical
data acquisition, health knowledge extraction, persistent data storage, adaptive visualization, system management and value-added services such as intelligent alerts, recommendations and guidelines.
mHealthDroid: a novel framework for agile development of mobile health applications
1. mHealthDroid:
a novel framework for agile
development of mobile health
applications
UCAmI & IWAAL 2014 (Belfast)
Oresti Baños, Rafael Garcia, Juan A. Holgado-Terriza, Miguel Damas,
Hector Pomares, Ignacio Rojas, Alejandro Saez and Claudia Villalonga
2. Context
• Technology has changed the healthcare paradigm
– Growing tendency in the use of mobile health applications
– Most of the apps are devoted to learning and formative
purposes
– User report VS wearable monitors
3. Context
• Mobile health is far from mature
– Scientists still need to build and validate mHealth
solutions
– mHealth apps focus on a special domain or lack of
essential features for health services
– Powerful frameworks and tools that support the
development are required A mHealth framework
4. Requirements of a mHealth Framework
• Provide rapid development
• Certain level of abstraction
– Support different devices
– Define a unified model
• Data storage and visualization
• Guidelines and Knowledge inference
– Signal processing
– Machine learning
– Intelligent recommendations
6. Architecture
• Communication Manager
– Abstraction
• Provides the abstraction level required to enable the
functioning of applications independently of the underlying
health technologies.
– Adapters
• Modules devised to support the use of an specific mobile or
biomedical device
• The Adapter manages the connection with the device,
interprets the received data and maps it to the unified data
model
– Extensible
• The modularity of the Adapters makes the Communication
Manager extensible and evolvable to future devices and
technologies.
7. Architecture
• Storage Manager
– Persistence
• Provides data persistence both locally and remotely abstracting
the queries from the underlying storage system
• Visualization Manager
– Online mode
• The data is provided by the Communication Manager at
runtime
– Offline mode
• The data is provided by the Storage Manager
8. Architecture
• Data Processing Manager
– Online mode
– Offline mode
– Modular
• The manager includes four independent modules typically used
in data processing.
9. Architecture
• Data Processing Manager
– Preprocessing
• This module is devised to apply mechanisms to clean,
transform and ultimately adequate the data to the specific
needs.
– Segmentation
• This module provides diverse techniques to split the data.
– Feature Extraction
• This module permits to transform the input data into a reduced
representation set of features or feature vector.
– Classification
• This module categorizes the data using the features extracted
by the Feature Extraction module
10. Architecture
• System Manager
– Provides functionalities to manage general resources of the mobile device
Wifi, GPS, Bluetooth, etc.
• Service Enablers
– Alerts Enabler
• Alerts procedures when abnormalities or risk situations are detected
– Notification Enabler
• Prescheduled or event-based user-friendly notifications
– Guidelines Enabler
• Multimedia tools for displaying personalized guidelines
– Medical Report Enabler
• Structuring the medical knowledge in an expert-oriented format
11. Data Model
• Must be
– Generic
– Flexible
– Extensible
Data • Data collected by the sensor
• Packages with the data from
all sensors Session
• Sample rate, start time, end
time
Session
Metadata
Sensors • Different supported sensors
12. mHealthDroid
• Android implementation of the mHealth framework
– Target to Android 4.2 but back compatibility from Android 2.3.3
– Released under the GNU GPLv3 license
– Source code github.com/mHealthTechnologies/mHealthDroid
• Communication Manager
– Provides adapters
for Android devices and
Shimmer devices
13. mHealthDroid
• Storage Manager
– SQLite for the local data management
– JSON for the transmission to a remote storage
• Visualization Manager
– External library for visualization (GraphView)
• Multiplot visualization
• Multisignal representation
• Graph customization
14. mHealthDroid
• Data Processing Manager
– Preprocessing
• Upsampling
• Downsampling
– Features Extraction
• Mean
• Variance
• Standard Deviation
• Zero and Mean Crossing Rate
• Maximum and Minimum
– Segmentation
• Sliding window
– Classification
• External library for machine
learning (WEKA) Naives
Bayes, Adaboost, Decisions
Tree, Linear Regression and
ZeroR
15. mHealthDroid
• System Manager
– Wifi
– Bluetooth
– Screen Brigthness
• Service Enablers
– Notifications
– Alerts
• Phone Calls
• Messages
– Guidelines
• Audio reproduction
• Video reproduction
• Youtube videos player
16. mHealthApp
• Exemplary app
– Composed by 6 tabs to illustrate the potential of mHealthDroid
– Available on Google Play
– Source code github.com/mHealthTechnologies/mHealthAPP
• Connectivity Tab
21. Conclusions
• mHealth is a very prominent field; however, there is a lack of tools
for the development of mHealth applications
• A novel mHealth framework which embraces the key requirements
of mHealth applications, namely, communication abstraction,
biomedical data acquisition, knowledge inference, data storage and
visualization, system management and services such as intelligent
alerts, recommendations and guidelines, is presented in this work
• mHealthDroid, an Android implementation of the mHealth
framework is described and made publicly available to the
community
• An application, particularly devoted to detect and track human
behavior, is developed to showcase the potential of mHealtDroid
22. Thank you for your attention.
Questions?
Alejandro Sáez Fernández
Master student at the Computer Technology Faculty of Computer
Science & Electrical Engineering (ETSIIT)
University of Granada, Granada (Spain)
Email: alejandrosaez3@gmail.com
Phone: +353 083 185 8701