Hassle Free Fitness Monitoring
description
Transcript of Hassle Free Fitness Monitoring
Hassle Free Fitness MonitoringHassle Free Fitness Monitoring
David Jea, Jason Liu, David Jea, Jason Liu,
Thomas Schmid, Mani SrivastavaThomas Schmid, Mani Srivastava
Pervasive Health Care SystemsPervasive Health Care Systems
Fitness Monitoring is the most Fundamental Functionality of Pervasive Fitness Monitoring is the most Fundamental Functionality of Pervasive Health Care SystemsHealth Care Systems
Provides 24X7 Fitness MonitoringProvides 24X7 Fitness Monitoring
Sensor devices are clipped on the bodySensor devices are clipped on the body
Proactively Record changes in vital signs such as weight and blood Proactively Record changes in vital signs such as weight and blood pressurepressure
Appropriate Medical Services provided on the basis of recorded dataAppropriate Medical Services provided on the basis of recorded data
ChallengesChallenges
PrivacyPrivacy
SecuritySecurity
Finding a perfect balance between usability, privacy and securityFinding a perfect balance between usability, privacy and security
ProblemsProblems
Large number of Devices hooked Large number of Devices hooked on the bodyon the body
Multiple type of sensorsMultiple type of sensors
Privacy concerns at workplacesPrivacy concerns at workplaces
Security IssuesSecurity Issues
Network Security IssuesNetwork Security Issues
User authentication issuesUser authentication issues
Security problems related to stolen Security problems related to stolen palmtops or PDA’spalmtops or PDA’s
The IdeaThe Idea
Build Fitness monitoring system for healthy individuals in a workplaceBuild Fitness monitoring system for healthy individuals in a workplace
Identification of the individual by only utilizing imprecise biometrics and Identification of the individual by only utilizing imprecise biometrics and existing informationexisting information
Maintaining the device’s original user interfaceMaintaining the device’s original user interface
No additional sensors incorporated in the systemNo additional sensors incorporated in the system
Design GuidelinesDesign Guidelines
PrivacyPrivacy Recorded data cannot be used as hard evidence (in court) to Recorded data cannot be used as hard evidence (in court) to
pinpoint exactly who the user ispinpoint exactly who the user is
FeasibilityFeasibility The system is allowed to use existing informationThe system is allowed to use existing information
UsabilityUsability Restoring the original interface of the device so that people of all age Restoring the original interface of the device so that people of all age
groups know how to use itgroups know how to use it
The DesignThe Design
Possible Candidates
Activity Information
Biometric Matcher
Context Reasoning
Imprecise Physiological
Info
Uncertainty Reduction
UserIdentity
ImplementationImplementation
The system consists of a weight scale and a blood pressure monitorThe system consists of a weight scale and a blood pressure monitor
Both devices communicate with the laptopBoth devices communicate with the laptop
Software program installed on laptop continuously record data and attach a Software program installed on laptop continuously record data and attach a timestamp to weight and blood pressure readingstimestamp to weight and blood pressure readings
Facility for a user to input his/her name is also providedFacility for a user to input his/her name is also provided This step is to establish ground truth for the experimentThis step is to establish ground truth for the experiment
Inference Engine ComponentsInference Engine Components
Biometric MatcherBiometric Matcher It implements a Bayes classifier that combines multiple sensor observationsIt implements a Bayes classifier that combines multiple sensor observations
It assumes that each observation is uniqueIt assumes that each observation is unique
This results in the identity of the subjectThis results in the identity of the subject
Context Reasoning Context Reasoning ItIt is based on Reified Temporal Logicis based on Reified Temporal Logic
It provides with the user’s contextIt provides with the user’s context
It uses two meta-Predicates to express when things are trueIt uses two meta-Predicates to express when things are true
AnalysisAnalysis
UserUser Similarity in Physiological Similarity in Physiological InformationInformation
Seat in LabSeat in Lab Usage HabitUsage Habit
Weight Weight ScaleScale
BP BP MonitorMonitor
BothBoth
AA LightLight VV VV
BB They have similar weights.They have similar weights.
The differences in mean The differences in mean are less than 1.9 lbsare less than 1.9 lbs
VV VV VV
CC VV VV
DD VV VV
EE Their Difference in Their Difference in average weights is 1.1 lbsaverage weights is 1.1 lbs
VV VV
FF VV
GG Their Difference in Their Difference in average weights is 1.1 lbsaverage weights is 1.1 lbs
VV VV
HH VV VV
II HeavyHeavy VV
Results for one Physiological InformationResults for one Physiological Information
Physiological Data for Classifier
Positive Match False Match
Weights 57.23% 45.7745.77
Systolic Blood Pressure 22.02% 77.9877.98
Diastolic Blood Pressure 43.90%43.90% 56.10%56.10%
Heartbeat Rate 25%25% 75%75%
Results based on Multiple Sources
Biometric matcher that combines all 4
physiological sources.
Positive
Match
False
Match
Classification Results for partial or complete
data points.
77.9% 22.1%
Classification Results for complete data
points only.
87.3% 12.7%
The accuracy of the context reasoning component
Context Reasoning Component Positive False Positive
The presence of a user based on
network activity
89.47% 10.53%
Combining the biometric matcher with the context
reasoning.
Biometric Matcher only Biometric Matcher and Context
Reasoning
Accuracy 78.16% 83.80%
ConclusionConclusion
Built a health monitoring system which is hassle freeBuilt a health monitoring system which is hassle free
Less privacy concernsLess privacy concerns
No extra sensors hooked on the bodyNo extra sensors hooked on the body
Easy to UseEasy to Use
Widely used by populationWidely used by population
How to handle uncertain usage?How to handle uncertain usage?