SlideShare ist ein Scribd-Unternehmen logo
1 von 4
Downloaden Sie, um offline zu lesen
A Mechanism for Detecting Trigger Points
and Irreversibility Thresholds in Shock and
      Trauma for Catastrophic Events
                                Martin DUDZIAK, PhD1
                           TETRAD Technologies Group, Inc., USA


           Abstract. We investigate the model of unstable recurrent patterns that emerge
           within chaotic and highly turbulent systems as a possible avenue for linking shifts
           in clinical parameters that can be indicators of nonlinear and irreversible
           transitions leading to mortality. The argument presented is that in the case of mass
           injury and trauma events from natural or intentional causes, large numbers of
           people may be in situations where clinical testing infrastructures have been
           disabled or destroyed, further reducing the ability of medical caregivers to
           accurately notice indicators and signals of impending critical and irreversible
           conditions. Identification of a reduced set of observables that can be linked with
           unstable yet recurrent patterns may provide a means for improved monitoring and
           life support under such adverse conditions.



           Keywords. shock, trauma, catastrophe, recurrent systems, anomaly detection,
           chaos, turbulence, dissipation, emergency preparedness, first response, triage



Introduction

Events such as natural disasters and deliberate terrorist-type attacks have the capability
of overtaxing even the most advanced systems for dealing with shock syndrome and
trauma which are generally tailored to the individual patient. Anomaly and intrusion
detection methods developed for network capacity planning, system configuration
management and information security, as well as the disciplines of fluid dynamics and
turbulence, appear to be distant from trauma medicine and the challenge of dealing
with mass trauma events such as from natural disasters, accidents or terrorist actions.
However, there may be a new class of parameters that can be measured rapidly for
large numbers of patients which can aid in the dual process of triage and forecast for
critical phases in emergency treatment of patients. Examination of unstable and semi-
stable recurrent patterns in dissipative extended systems may offer a set of advance-
warning indicators that can be associated with impending critical life support

  1
    Corresponding author: TETRAD Technologies Group, Inc., 28 Chase Gayton Circle, Suite 736,
Richmond VA 23238 USA Email: martin@forteplan.com
conditions, in some cases clearly irreversible, in others treatable if addressed
sufficiently early in time. The target situation is that which occurs when there are
hundreds or even thousands of persons subjected, within a comparatively short period
of time (minutes or hours) by exposure or ingestion to toxic biological agents,
chemicals, or radiation, or alternatively subjected to injuries from impact events (e.g.,
shock waves, blunt trauma).

     Our goal in this brief speculative note is to set the stage for inquiry and
investigation first at an abstract level and then to identify if the abstract model that
results can find a match with known observable parameters from the critical care world,
working abductively and intuitively toward defining a dataset that can in turn be tested
for a fit with both the turbulence characteristics and the existence of unstable and semi-
stable recurrent patterns (hereafter referred to as λ-recurrence) fitting the initial abstract
model. No experiments have yet been conducted and the present work reflects early
stage simulation and modeling.

     We begin by considering the complex space of biochemical and behavioral
parameters that are observable and with constraints as described earlier. Thus,
automatically we must rule out any medical testing that requires remote and time-
intensive laboratory testing that cannot be performed in an emergency room “live”
without sending samples or patients to other facilities. In fact we prefer to focus upon
those parameters that can be collected reliably in a field setting such as that of a combat
area field hospital or onboard a medivac helicopter. Certainly within the space of
possible parameters are those employed in the Glasgow Outcome Scale but from the
outset we do not know if we should be considering one particular parameter p (e.g.,
SpO2/SaO2) or a set S = {p, q, r}, for which the relationship between two or more
members of S is the critical meta-parameter; i.e, where we may find chaotic behavior of
the “interesting” sort (meaning, with λ-recurrence).

     Given the constraints of our interest in mass trauma events coupled with disruption
of basic infrastructure and resources, we are automatically limited in the range of
clinical tests that can be considered, although as technology evolves, there may be new
non-invasive or rapid and portable analyses opening up other dimensions. Parameter
sets useful for tracking λ-recurrence may emerge including quantities that are relatively
stable on scales of minutes or even hours but for which modest changes coupled with
other set members can serve as triggers of a λ-recurrence. Measures of pro- and anti-
inflammatory cytokine and adhesion molecule patterns [1] are already employed on a 2
to 4 hour basis for indications of septic shock onset or susceptibility. Within our
investigations, the In-Check microfluidic diagnostic chip and platform [2] offers the
possibility of such new classes of tests through highly-automated PCR-based analysis
in a portable lightweight and low-power configuration capable of performing up to five
tests and protocols in parallel.


Recurrent Patterns in Dissipative Turbulence - the Kuramoto-Sivashinsky Model

Our goal is to find behaviors similar to the basic Kuramoto-Sivashinsky (KS) system,

      ut = (u2)x – uxx - vuxxxx                                                            (1)
representing, for example, the general case for amplitudes of interfacial instabilities
such as flame fields, nonlinear through term (u2)x, t referring to time and v a damping
parameter. The solution u(x,t) = u(x+L, t) is periodic and can be expanded into a
discrete spatial Fourier series
                  +∞
  u ( x, t ) =   ∑a
                 k = −∞
                          k   (t )e ikx                                              (2)


    It is possible to extend periodic orbit theory to spatiotemporal chaotic systems and
the KS system [3] and reduce a high-demensional system (from 16 to 64 dimensions)
down to one, namely a return map. Following the seminal work of Cvitanović and
others [4,5] we postulate that a predictive system for non-equilibrium turbulence of the
type that may exist in breathing irregularities, cardiac beat-to-beat chaos and other
chaotic-like behavior measured through electrocardiograms, or in molecular pattern2s
such as cytokine levels, may require only 102 – 104 recurrent pattern instances as
opposed to classical expectations (≈ 1010) from Monte Carlo and other PDE simulations.
This means fewer cycles of per-patient monitoring and possibly a manageable cycle of
acquire-search-compare-evaluate computations to perform (returning to our clinical
focus) for large populations when situations impose sparser data collection intervals
and less robust computational resources.

     Dynamical behavior such as 3-D Navier-Stokes turbulence in many different
media often exhibits a variety of unstable recurrent patterns. Such recurrent behavior
can in turn be applied to provide indicators of general characteristics or imminent phase
shifts. The challenge is how to find such patterns since by definition they are
themselves unstable and not the same in amplitude or other dynamical characteristics in
each recurrence. Why they may be indicators of trigger-like conditions, including
irreversible conditions leading to death, in the case of individuals experiencing a
variety of trauma conditions is not at all obvious. However, on the one hand there is
the evidence of sudden death linked with a larger variety of CNS and cardiovascular
conditions for which there are no externally observable nor commonly measureable
(e.g., pulse, pressure, breath rate, consciousness) indicators but which have may have
increases or decreases in chaotic patterns such as EKG.


How to find and associate recurrent patterns with biomedical conditions

Finding periodic orbits may be enhanced by algorithms but it is still an exploratory and
somewhat “trial and error” process. More so is the conjectured association with
medical conditions. Recent studies by Cvitanović and Lan involve an iterative
approach to begin with a calculated guess and then to monitor the attraction of the
solution towards or away from this target. This in turns builds an approach using
Newton-Raphson or alternatively by minimizing a cost function related to the
difference between approximate flow and discrete flow of that which is circulating or

    2
      It is important to remember that recurrent patterns of interest are not necessarily
themselves chaotic and in fact the absence or decrease of chaos in certain
bioparameters may be the indicator of on onset of a disabling or life-threatening
condition, as pointed out by Poon and Merrill regarding congestive heart failure [6]
has a potential for doing so actively. These and other methods like them will not tell
which patterns are recurrent nor how “near” everything is in the interpolative process.
Here additional heuristics and probabilistic data can be used to reduce the iterative
cycle. There are known techniques within the study of anomaly detection and tracking,
such as multivariate statistics and neural network classification. At a higher scale
probabilistic (Bayesian) networks can be employed with specific rules drawn from the
operations domain.

    Once there is a library of unstable spatio-temporally recurrent solutions there must
be some form of application in this case to a type or configuration of biometric
parameters. Periodic orbit theory [7] proposes that the greater the unstability, the more
accurate will be the predictions that are based upon a small number of the shortest
recurrent patterns. In turn, once a sufficient number of individual unique patterns can
be discriminated, then these can be use together to predict global averages. Average is
considered as a sum over all possible patterns which are grouped hierarchically
according to the likelihood of each pattern’s occrrence in the system.
                  1 t                t
      a = lim       A , A t ( x ) = ∫ dτa ( x (τ ))                                (3)
             t →∞ t                  0

where the dynamical trajectory x(t) is some point in a high-dimension state space and
the object is for short cycles, short patterns, to contribte information collectively as an
aggregate to information on the invariant set, the long-term cycle. Returning to the
shock-trauma context of individuals potentially injured internally and unconscious or
delirious and unable to assist informationally, with medical caregivers overburdened,
with resources and infrastructures stretched or non-operable, the idea here is to have a
way to make a logical step from simple observable patterns that can be captured
relatively easily, through pulse oxymetry, temperature, pressure, pulse, and other
means, to information about impending, imminent shifts in the long-term cycle, the :big
picture” of systemic changes that could be the difference between death and life.


References

[1] V von Dossow, K Rotard, U Redlich, O V Hein and CD Spies, Circulating immune
parameters predicting the progression from hospital-acquired pneumonia to septic
shock in surgical patients, Critical Care 2005, 9:R662-R669
[2] White papers and other technical data on the In-Check platform are available from
ST Microelectronics http://www.st.com/stonline/prodpres/dedicate/labchip/labchip.htm
[3] P Cvitanović and T Lan, Turbulent fields and their recurrences, in N Antoniou, ed.,
Proc. Of 10th Int’l Workshop on Multiparticle Production: Correlations and
Fluctuations in QCD, World Scientific, Singapore 2003
[4] E Hopf, A mathematical example displaying features of turbulence, Commun. of
Applied Math. 1, 303 (1948)
[5] P Holmes, JL Lumley and G Berkooz, Turbulence, Coherent Structures, Dynamical
Systems and Symmetry, Cambridge Univ. Press, Cambridge 1996
[6] CS Poon and CK Merrill, Decrease of cardiac chaos in congestive heart failure,
Nature 1997 Oct 2;389(6650):492-5
[7] P Cvitanović et al, Chaos: Quantum and Classical, Niels Bohr Institute,
Copenhagen 2003

Weitere ähnliche Inhalte

Andere mochten auch

Confrontan los resultados de esta jornada con los obtenidos en la primera y r...
Confrontan los resultados de esta jornada con los obtenidos en la primera y r...Confrontan los resultados de esta jornada con los obtenidos en la primera y r...
Confrontan los resultados de esta jornada con los obtenidos en la primera y r...Ale Cosali
 
Systems Assessment - Workshop Framework
Systems Assessment - Workshop FrameworkSystems Assessment - Workshop Framework
Systems Assessment - Workshop FrameworkJeff Granger
 
урок математики
урок математикиурок математики
урок математикиOksanaPe
 
Press Release 1 T04 Tele Celular Sul
Press Release 1 T04   Tele Celular SulPress Release 1 T04   Tele Celular Sul
Press Release 1 T04 Tele Celular SulTIM RI
 
хүнснйи ногоо
хүнснйи ногоохүнснйи ногоо
хүнснйи ногооarunaa-53
 
династия романовых
династия романовыхдинастия романовых
династия романовыхnzayceva
 
A 08-suffering- part i-sanctified walking
A 08-suffering- part i-sanctified walkingA 08-suffering- part i-sanctified walking
A 08-suffering- part i-sanctified walkingCybermissions
 
Integrating bpi and innovation
Integrating bpi and innovationIntegrating bpi and innovation
Integrating bpi and innovationDoug Brockway
 
Wicentowski - XML for the Rest of Us
Wicentowski - XML for the Rest of UsWicentowski - XML for the Rest of Us
Wicentowski - XML for the Rest of Usjbflint
 
Jessica Ennis's Gallery
Jessica Ennis's GalleryJessica Ennis's Gallery
Jessica Ennis's Galleryemmarob12393
 

Andere mochten auch (17)

Confrontan los resultados de esta jornada con los obtenidos en la primera y r...
Confrontan los resultados de esta jornada con los obtenidos en la primera y r...Confrontan los resultados de esta jornada con los obtenidos en la primera y r...
Confrontan los resultados de esta jornada con los obtenidos en la primera y r...
 
Smart board cuidados
Smart board   cuidadosSmart board   cuidados
Smart board cuidados
 
Systems Assessment - Workshop Framework
Systems Assessment - Workshop FrameworkSystems Assessment - Workshop Framework
Systems Assessment - Workshop Framework
 
урок математики
урок математикиурок математики
урок математики
 
Press Release 1 T04 Tele Celular Sul
Press Release 1 T04   Tele Celular SulPress Release 1 T04   Tele Celular Sul
Press Release 1 T04 Tele Celular Sul
 
Yuri y su tecnologia
Yuri y su tecnologiaYuri y su tecnologia
Yuri y su tecnologia
 
Fajardo
FajardoFajardo
Fajardo
 
Sense Net Mmvr2005
Sense Net Mmvr2005Sense Net Mmvr2005
Sense Net Mmvr2005
 
хүнснйи ногоо
хүнснйи ногоохүнснйи ногоо
хүнснйи ногоо
 
династия романовых
династия романовыхдинастия романовых
династия романовых
 
Atletismo inés5
Atletismo inés5Atletismo inés5
Atletismo inés5
 
A 08-suffering- part i-sanctified walking
A 08-suffering- part i-sanctified walkingA 08-suffering- part i-sanctified walking
A 08-suffering- part i-sanctified walking
 
илтгэл
илтгэлилтгэл
илтгэл
 
Integrating bpi and innovation
Integrating bpi and innovationIntegrating bpi and innovation
Integrating bpi and innovation
 
Tulhuuriin juram
Tulhuuriin juramTulhuuriin juram
Tulhuuriin juram
 
Wicentowski - XML for the Rest of Us
Wicentowski - XML for the Rest of UsWicentowski - XML for the Rest of Us
Wicentowski - XML for the Rest of Us
 
Jessica Ennis's Gallery
Jessica Ennis's GalleryJessica Ennis's Gallery
Jessica Ennis's Gallery
 

Ähnlich wie Shocktraumamodel Mmvr2005

A Mechanistic, Stochastic Model Helps Understand Multiple Sclerosis Course an...
A Mechanistic, Stochastic Model Helps Understand Multiple Sclerosis Course an...A Mechanistic, Stochastic Model Helps Understand Multiple Sclerosis Course an...
A Mechanistic, Stochastic Model Helps Understand Multiple Sclerosis Course an...Mutiple Sclerosis
 
Nonlinear methods of analysis of electrophysiological data and Machine learni...
Nonlinear methods of analysis of electrophysiological data and Machine learni...Nonlinear methods of analysis of electrophysiological data and Machine learni...
Nonlinear methods of analysis of electrophysiological data and Machine learni...Facultad de Informática UCM
 
Project Description Of A Mathematical Model
Project Description Of A Mathematical ModelProject Description Of A Mathematical Model
Project Description Of A Mathematical ModelAnn Johnson
 
Logistic Loglogistic With Long Term Survivors For Split Population Model
Logistic Loglogistic With Long Term Survivors For Split Population ModelLogistic Loglogistic With Long Term Survivors For Split Population Model
Logistic Loglogistic With Long Term Survivors For Split Population ModelWaqas Tariq
 
Markov chain and SIR epidemic model (Greenwood model)
Markov chain and SIR epidemic model (Greenwood model)Markov chain and SIR epidemic model (Greenwood model)
Markov chain and SIR epidemic model (Greenwood model)writwik mandal
 
Rodriguez_Ullmayer_Rojo_RUSIS@UNR_REU_Technical_Report
Rodriguez_Ullmayer_Rojo_RUSIS@UNR_REU_Technical_ReportRodriguez_Ullmayer_Rojo_RUSIS@UNR_REU_Technical_Report
Rodriguez_Ullmayer_Rojo_RUSIS@UNR_REU_Technical_Report​Iván Rodríguez
 
ICAF 2007 - On the Selection of Test Factors for the Determination of Safe Life
ICAF 2007 - On the Selection of Test Factors for the Determination of Safe LifeICAF 2007 - On the Selection of Test Factors for the Determination of Safe Life
ICAF 2007 - On the Selection of Test Factors for the Determination of Safe LifeGuy Habermann (MBA BSc BEng Hons 1)
 
Probability Models for Estimating Haplotype Frequencies and Bayesian Survival...
Probability Models for Estimating Haplotype Frequencies and Bayesian Survival...Probability Models for Estimating Haplotype Frequencies and Bayesian Survival...
Probability Models for Estimating Haplotype Frequencies and Bayesian Survival...Université de Dschang
 
Amodelbasedoncellularautomatatosimulatea si sepidemicdisease
Amodelbasedoncellularautomatatosimulatea si sepidemicdiseaseAmodelbasedoncellularautomatatosimulatea si sepidemicdisease
Amodelbasedoncellularautomatatosimulatea si sepidemicdiseaseCarlosReyes671
 
A Wearable Accelerometer System for Unobtrusive Monitoring of Parkinson’s Dis...
A Wearable Accelerometer System for Unobtrusive Monitoring of Parkinson’s Dis...A Wearable Accelerometer System for Unobtrusive Monitoring of Parkinson’s Dis...
A Wearable Accelerometer System for Unobtrusive Monitoring of Parkinson’s Dis...Michael J. Montgomery
 
Use Proportional Hazards Regression Method To Analyze The Survival of Patient...
Use Proportional Hazards Regression Method To Analyze The Survival of Patient...Use Proportional Hazards Regression Method To Analyze The Survival of Patient...
Use Proportional Hazards Regression Method To Analyze The Survival of Patient...Waqas Tariq
 
Improving predictability and performance by relating the number of events and...
Improving predictability and performance by relating the number of events and...Improving predictability and performance by relating the number of events and...
Improving predictability and performance by relating the number of events and...Asoka Korale
 

Ähnlich wie Shocktraumamodel Mmvr2005 (20)

Survival analysis
Survival analysisSurvival analysis
Survival analysis
 
A Mechanistic, Stochastic Model Helps Understand Multiple Sclerosis Course an...
A Mechanistic, Stochastic Model Helps Understand Multiple Sclerosis Course an...A Mechanistic, Stochastic Model Helps Understand Multiple Sclerosis Course an...
A Mechanistic, Stochastic Model Helps Understand Multiple Sclerosis Course an...
 
Nonlinear methods of analysis of electrophysiological data and Machine learni...
Nonlinear methods of analysis of electrophysiological data and Machine learni...Nonlinear methods of analysis of electrophysiological data and Machine learni...
Nonlinear methods of analysis of electrophysiological data and Machine learni...
 
Project Description Of A Mathematical Model
Project Description Of A Mathematical ModelProject Description Of A Mathematical Model
Project Description Of A Mathematical Model
 
Part 1 Survival Analysis
Part 1 Survival AnalysisPart 1 Survival Analysis
Part 1 Survival Analysis
 
TestSurvRec manual
TestSurvRec manualTestSurvRec manual
TestSurvRec manual
 
Logistic Loglogistic With Long Term Survivors For Split Population Model
Logistic Loglogistic With Long Term Survivors For Split Population ModelLogistic Loglogistic With Long Term Survivors For Split Population Model
Logistic Loglogistic With Long Term Survivors For Split Population Model
 
Markov chain and SIR epidemic model (Greenwood model)
Markov chain and SIR epidemic model (Greenwood model)Markov chain and SIR epidemic model (Greenwood model)
Markov chain and SIR epidemic model (Greenwood model)
 
Cox model
Cox modelCox model
Cox model
 
Rodriguez_Ullmayer_Rojo_RUSIS@UNR_REU_Technical_Report
Rodriguez_Ullmayer_Rojo_RUSIS@UNR_REU_Technical_ReportRodriguez_Ullmayer_Rojo_RUSIS@UNR_REU_Technical_Report
Rodriguez_Ullmayer_Rojo_RUSIS@UNR_REU_Technical_Report
 
ICAF 2007 - On the Selection of Test Factors for the Determination of Safe Life
ICAF 2007 - On the Selection of Test Factors for the Determination of Safe LifeICAF 2007 - On the Selection of Test Factors for the Determination of Safe Life
ICAF 2007 - On the Selection of Test Factors for the Determination of Safe Life
 
Art01
Art01Art01
Art01
 
Part 2 Cox Regression
Part 2 Cox RegressionPart 2 Cox Regression
Part 2 Cox Regression
 
Probability Models for Estimating Haplotype Frequencies and Bayesian Survival...
Probability Models for Estimating Haplotype Frequencies and Bayesian Survival...Probability Models for Estimating Haplotype Frequencies and Bayesian Survival...
Probability Models for Estimating Haplotype Frequencies and Bayesian Survival...
 
Amodelbasedoncellularautomatatosimulatea si sepidemicdisease
Amodelbasedoncellularautomatatosimulatea si sepidemicdiseaseAmodelbasedoncellularautomatatosimulatea si sepidemicdisease
Amodelbasedoncellularautomatatosimulatea si sepidemicdisease
 
A Wearable Accelerometer System for Unobtrusive Monitoring of Parkinson’s Dis...
A Wearable Accelerometer System for Unobtrusive Monitoring of Parkinson’s Dis...A Wearable Accelerometer System for Unobtrusive Monitoring of Parkinson’s Dis...
A Wearable Accelerometer System for Unobtrusive Monitoring of Parkinson’s Dis...
 
Use Proportional Hazards Regression Method To Analyze The Survival of Patient...
Use Proportional Hazards Regression Method To Analyze The Survival of Patient...Use Proportional Hazards Regression Method To Analyze The Survival of Patient...
Use Proportional Hazards Regression Method To Analyze The Survival of Patient...
 
Af4101182186
Af4101182186Af4101182186
Af4101182186
 
Scheffer
SchefferScheffer
Scheffer
 
Improving predictability and performance by relating the number of events and...
Improving predictability and performance by relating the number of events and...Improving predictability and performance by relating the number of events and...
Improving predictability and performance by relating the number of events and...
 

Mehr von martindudziak

Annot Mjd Selected Present Course Train Aug07
Annot Mjd Selected Present Course Train Aug07Annot Mjd Selected Present Course Train Aug07
Annot Mjd Selected Present Course Train Aug07martindudziak
 
Coordinated And Unified Responses To Unpredictable And Widespread Biothreats
Coordinated And Unified Responses To Unpredictable And Widespread BiothreatsCoordinated And Unified Responses To Unpredictable And Widespread Biothreats
Coordinated And Unified Responses To Unpredictable And Widespread Biothreatsmartindudziak
 
New Iaa Magopt Paper
New Iaa Magopt PaperNew Iaa Magopt Paper
New Iaa Magopt Papermartindudziak
 
Redbionet Idhs White Paper
Redbionet Idhs White PaperRedbionet Idhs White Paper
Redbionet Idhs White Papermartindudziak
 
Radterrorism Spb Oct04
Radterrorism Spb Oct04Radterrorism Spb Oct04
Radterrorism Spb Oct04martindudziak
 
Radterror Spb Oct04 Paper
Radterror Spb Oct04 PaperRadterror Spb Oct04 Paper
Radterror Spb Oct04 Papermartindudziak
 
Empires Idhs Red Cell White Paper
Empires Idhs Red Cell White PaperEmpires Idhs Red Cell White Paper
Empires Idhs Red Cell White Papermartindudziak
 
Berlin Intl Sec Conf2 Paper
Berlin Intl Sec Conf2 PaperBerlin Intl Sec Conf2 Paper
Berlin Intl Sec Conf2 Papermartindudziak
 
Evolutionary IED Prevention 09 2006 Updated W Comments Jan2010
Evolutionary IED Prevention 09 2006 Updated W Comments Jan2010Evolutionary IED Prevention 09 2006 Updated W Comments Jan2010
Evolutionary IED Prevention 09 2006 Updated W Comments Jan2010martindudziak
 

Mehr von martindudziak (20)

Annot Mjd Selected Present Course Train Aug07
Annot Mjd Selected Present Course Train Aug07Annot Mjd Selected Present Course Train Aug07
Annot Mjd Selected Present Course Train Aug07
 
Coordinated And Unified Responses To Unpredictable And Widespread Biothreats
Coordinated And Unified Responses To Unpredictable And Widespread BiothreatsCoordinated And Unified Responses To Unpredictable And Widespread Biothreats
Coordinated And Unified Responses To Unpredictable And Widespread Biothreats
 
Structinteg
StructintegStructinteg
Structinteg
 
Ndemetalassemb
NdemetalassembNdemetalassemb
Ndemetalassemb
 
Nanosp98paper
Nanosp98paperNanosp98paper
Nanosp98paper
 
New Iaa Magopt Paper
New Iaa Magopt PaperNew Iaa Magopt Paper
New Iaa Magopt Paper
 
Spmnanopres
SpmnanopresSpmnanopres
Spmnanopres
 
Landmines
LandminesLandmines
Landmines
 
Ihee Ppres0998
Ihee Ppres0998Ihee Ppres0998
Ihee Ppres0998
 
Redbionet Idhs White Paper
Redbionet Idhs White PaperRedbionet Idhs White Paper
Redbionet Idhs White Paper
 
Radterrorism Spb Oct04
Radterrorism Spb Oct04Radterrorism Spb Oct04
Radterrorism Spb Oct04
 
Radterror Spb Oct04 Paper
Radterror Spb Oct04 PaperRadterror Spb Oct04 Paper
Radterror Spb Oct04 Paper
 
Empires Idhs Red Cell White Paper
Empires Idhs Red Cell White PaperEmpires Idhs Red Cell White Paper
Empires Idhs Red Cell White Paper
 
Berlin Intl Sec Conf2 Paper
Berlin Intl Sec Conf2 PaperBerlin Intl Sec Conf2 Paper
Berlin Intl Sec Conf2 Paper
 
Ecosymbiotics01
Ecosymbiotics01Ecosymbiotics01
Ecosymbiotics01
 
Qnsrbio
QnsrbioQnsrbio
Qnsrbio
 
Tdbasicsem
TdbasicsemTdbasicsem
Tdbasicsem
 
Bistablecamnets
BistablecamnetsBistablecamnets
Bistablecamnets
 
Sciconsc
SciconscSciconsc
Sciconsc
 
Evolutionary IED Prevention 09 2006 Updated W Comments Jan2010
Evolutionary IED Prevention 09 2006 Updated W Comments Jan2010Evolutionary IED Prevention 09 2006 Updated W Comments Jan2010
Evolutionary IED Prevention 09 2006 Updated W Comments Jan2010
 

Shocktraumamodel Mmvr2005

  • 1. A Mechanism for Detecting Trigger Points and Irreversibility Thresholds in Shock and Trauma for Catastrophic Events Martin DUDZIAK, PhD1 TETRAD Technologies Group, Inc., USA Abstract. We investigate the model of unstable recurrent patterns that emerge within chaotic and highly turbulent systems as a possible avenue for linking shifts in clinical parameters that can be indicators of nonlinear and irreversible transitions leading to mortality. The argument presented is that in the case of mass injury and trauma events from natural or intentional causes, large numbers of people may be in situations where clinical testing infrastructures have been disabled or destroyed, further reducing the ability of medical caregivers to accurately notice indicators and signals of impending critical and irreversible conditions. Identification of a reduced set of observables that can be linked with unstable yet recurrent patterns may provide a means for improved monitoring and life support under such adverse conditions. Keywords. shock, trauma, catastrophe, recurrent systems, anomaly detection, chaos, turbulence, dissipation, emergency preparedness, first response, triage Introduction Events such as natural disasters and deliberate terrorist-type attacks have the capability of overtaxing even the most advanced systems for dealing with shock syndrome and trauma which are generally tailored to the individual patient. Anomaly and intrusion detection methods developed for network capacity planning, system configuration management and information security, as well as the disciplines of fluid dynamics and turbulence, appear to be distant from trauma medicine and the challenge of dealing with mass trauma events such as from natural disasters, accidents or terrorist actions. However, there may be a new class of parameters that can be measured rapidly for large numbers of patients which can aid in the dual process of triage and forecast for critical phases in emergency treatment of patients. Examination of unstable and semi- stable recurrent patterns in dissipative extended systems may offer a set of advance- warning indicators that can be associated with impending critical life support 1 Corresponding author: TETRAD Technologies Group, Inc., 28 Chase Gayton Circle, Suite 736, Richmond VA 23238 USA Email: martin@forteplan.com
  • 2. conditions, in some cases clearly irreversible, in others treatable if addressed sufficiently early in time. The target situation is that which occurs when there are hundreds or even thousands of persons subjected, within a comparatively short period of time (minutes or hours) by exposure or ingestion to toxic biological agents, chemicals, or radiation, or alternatively subjected to injuries from impact events (e.g., shock waves, blunt trauma). Our goal in this brief speculative note is to set the stage for inquiry and investigation first at an abstract level and then to identify if the abstract model that results can find a match with known observable parameters from the critical care world, working abductively and intuitively toward defining a dataset that can in turn be tested for a fit with both the turbulence characteristics and the existence of unstable and semi- stable recurrent patterns (hereafter referred to as λ-recurrence) fitting the initial abstract model. No experiments have yet been conducted and the present work reflects early stage simulation and modeling. We begin by considering the complex space of biochemical and behavioral parameters that are observable and with constraints as described earlier. Thus, automatically we must rule out any medical testing that requires remote and time- intensive laboratory testing that cannot be performed in an emergency room “live” without sending samples or patients to other facilities. In fact we prefer to focus upon those parameters that can be collected reliably in a field setting such as that of a combat area field hospital or onboard a medivac helicopter. Certainly within the space of possible parameters are those employed in the Glasgow Outcome Scale but from the outset we do not know if we should be considering one particular parameter p (e.g., SpO2/SaO2) or a set S = {p, q, r}, for which the relationship between two or more members of S is the critical meta-parameter; i.e, where we may find chaotic behavior of the “interesting” sort (meaning, with λ-recurrence). Given the constraints of our interest in mass trauma events coupled with disruption of basic infrastructure and resources, we are automatically limited in the range of clinical tests that can be considered, although as technology evolves, there may be new non-invasive or rapid and portable analyses opening up other dimensions. Parameter sets useful for tracking λ-recurrence may emerge including quantities that are relatively stable on scales of minutes or even hours but for which modest changes coupled with other set members can serve as triggers of a λ-recurrence. Measures of pro- and anti- inflammatory cytokine and adhesion molecule patterns [1] are already employed on a 2 to 4 hour basis for indications of septic shock onset or susceptibility. Within our investigations, the In-Check microfluidic diagnostic chip and platform [2] offers the possibility of such new classes of tests through highly-automated PCR-based analysis in a portable lightweight and low-power configuration capable of performing up to five tests and protocols in parallel. Recurrent Patterns in Dissipative Turbulence - the Kuramoto-Sivashinsky Model Our goal is to find behaviors similar to the basic Kuramoto-Sivashinsky (KS) system, ut = (u2)x – uxx - vuxxxx (1)
  • 3. representing, for example, the general case for amplitudes of interfacial instabilities such as flame fields, nonlinear through term (u2)x, t referring to time and v a damping parameter. The solution u(x,t) = u(x+L, t) is periodic and can be expanded into a discrete spatial Fourier series +∞ u ( x, t ) = ∑a k = −∞ k (t )e ikx (2) It is possible to extend periodic orbit theory to spatiotemporal chaotic systems and the KS system [3] and reduce a high-demensional system (from 16 to 64 dimensions) down to one, namely a return map. Following the seminal work of Cvitanović and others [4,5] we postulate that a predictive system for non-equilibrium turbulence of the type that may exist in breathing irregularities, cardiac beat-to-beat chaos and other chaotic-like behavior measured through electrocardiograms, or in molecular pattern2s such as cytokine levels, may require only 102 – 104 recurrent pattern instances as opposed to classical expectations (≈ 1010) from Monte Carlo and other PDE simulations. This means fewer cycles of per-patient monitoring and possibly a manageable cycle of acquire-search-compare-evaluate computations to perform (returning to our clinical focus) for large populations when situations impose sparser data collection intervals and less robust computational resources. Dynamical behavior such as 3-D Navier-Stokes turbulence in many different media often exhibits a variety of unstable recurrent patterns. Such recurrent behavior can in turn be applied to provide indicators of general characteristics or imminent phase shifts. The challenge is how to find such patterns since by definition they are themselves unstable and not the same in amplitude or other dynamical characteristics in each recurrence. Why they may be indicators of trigger-like conditions, including irreversible conditions leading to death, in the case of individuals experiencing a variety of trauma conditions is not at all obvious. However, on the one hand there is the evidence of sudden death linked with a larger variety of CNS and cardiovascular conditions for which there are no externally observable nor commonly measureable (e.g., pulse, pressure, breath rate, consciousness) indicators but which have may have increases or decreases in chaotic patterns such as EKG. How to find and associate recurrent patterns with biomedical conditions Finding periodic orbits may be enhanced by algorithms but it is still an exploratory and somewhat “trial and error” process. More so is the conjectured association with medical conditions. Recent studies by Cvitanović and Lan involve an iterative approach to begin with a calculated guess and then to monitor the attraction of the solution towards or away from this target. This in turns builds an approach using Newton-Raphson or alternatively by minimizing a cost function related to the difference between approximate flow and discrete flow of that which is circulating or 2 It is important to remember that recurrent patterns of interest are not necessarily themselves chaotic and in fact the absence or decrease of chaos in certain bioparameters may be the indicator of on onset of a disabling or life-threatening condition, as pointed out by Poon and Merrill regarding congestive heart failure [6]
  • 4. has a potential for doing so actively. These and other methods like them will not tell which patterns are recurrent nor how “near” everything is in the interpolative process. Here additional heuristics and probabilistic data can be used to reduce the iterative cycle. There are known techniques within the study of anomaly detection and tracking, such as multivariate statistics and neural network classification. At a higher scale probabilistic (Bayesian) networks can be employed with specific rules drawn from the operations domain. Once there is a library of unstable spatio-temporally recurrent solutions there must be some form of application in this case to a type or configuration of biometric parameters. Periodic orbit theory [7] proposes that the greater the unstability, the more accurate will be the predictions that are based upon a small number of the shortest recurrent patterns. In turn, once a sufficient number of individual unique patterns can be discriminated, then these can be use together to predict global averages. Average is considered as a sum over all possible patterns which are grouped hierarchically according to the likelihood of each pattern’s occrrence in the system. 1 t t a = lim A , A t ( x ) = ∫ dτa ( x (τ )) (3) t →∞ t 0 where the dynamical trajectory x(t) is some point in a high-dimension state space and the object is for short cycles, short patterns, to contribte information collectively as an aggregate to information on the invariant set, the long-term cycle. Returning to the shock-trauma context of individuals potentially injured internally and unconscious or delirious and unable to assist informationally, with medical caregivers overburdened, with resources and infrastructures stretched or non-operable, the idea here is to have a way to make a logical step from simple observable patterns that can be captured relatively easily, through pulse oxymetry, temperature, pressure, pulse, and other means, to information about impending, imminent shifts in the long-term cycle, the :big picture” of systemic changes that could be the difference between death and life. References [1] V von Dossow, K Rotard, U Redlich, O V Hein and CD Spies, Circulating immune parameters predicting the progression from hospital-acquired pneumonia to septic shock in surgical patients, Critical Care 2005, 9:R662-R669 [2] White papers and other technical data on the In-Check platform are available from ST Microelectronics http://www.st.com/stonline/prodpres/dedicate/labchip/labchip.htm [3] P Cvitanović and T Lan, Turbulent fields and their recurrences, in N Antoniou, ed., Proc. Of 10th Int’l Workshop on Multiparticle Production: Correlations and Fluctuations in QCD, World Scientific, Singapore 2003 [4] E Hopf, A mathematical example displaying features of turbulence, Commun. of Applied Math. 1, 303 (1948) [5] P Holmes, JL Lumley and G Berkooz, Turbulence, Coherent Structures, Dynamical Systems and Symmetry, Cambridge Univ. Press, Cambridge 1996 [6] CS Poon and CK Merrill, Decrease of cardiac chaos in congestive heart failure, Nature 1997 Oct 2;389(6650):492-5 [7] P Cvitanović et al, Chaos: Quantum and Classical, Niels Bohr Institute, Copenhagen 2003