Welcome to the Aging Services Technologies Laboratory
The Aging Services Technologies Laboratory is an interdisciplinary research lab focused on developing innovative technology and systems that improve elderly people’s quality of life.
Our vision is to create the technological foundation for maintaining a high quality of life as people age.
Our mission is to develop technology that will:
Increase quality of lifeDecrease health-care costsBe applicable to our soldiers, veterans and people with disabilities as well
Partnership Opportunities Partnership opportunities are available for companies, institutions and investors.
QuickTime™ and a decompressor
are needed to see this picture.
QoLT Lab QoLT Lab
increasingincreasing youryour QoLQoL
VisionVision
To become a global leader in performing research and innovating technologies in increasing the quality of life with aging.
To become a global leader in performing research and innovating technologies in increasing the quality of life with aging.
QoLT Lab QoLT Lab
increasingincreasing youryour QoLQoL
TeamTeamTeamTeam• Lakshman Tamil, Electrical Engineering
• Architecture, radio and overall management of the project
• Subhash Banerjee, M.D., UTSW Medical Center• Cardiology
• Gopal Gupta, Computer Science• Software
• Larry Amman, Mathematics & Statistics• Statistical analysis & Modeling
• Mehrdad Nourani, Electrical Engineering• Hardware, integration & testing, ASIC/SoC design
• Hlaing Minn, Electrical Engineering• Communication hardware design and modeling
• Vincent Ng, Computer Science• Machine learning
• Lakshman Tamil, Electrical Engineering• Architecture, radio and overall management of the project
• Subhash Banerjee, M.D., UTSW Medical Center• Cardiology
• Gopal Gupta, Computer Science• Software
• Larry Amman, Mathematics & Statistics• Statistical analysis & Modeling
• Mehrdad Nourani, Electrical Engineering• Hardware, integration & testing, ASIC/SoC design
• Hlaing Minn, Electrical Engineering• Communication hardware design and modeling
• Vincent Ng, Computer Science• Machine learning
Generic Body Area Network (BAN)
Generic Body Area Network (BAN)
• Several non-invasive sensors worn on body• Vital signs data collected and passed (via
gateway) to system database• Database stores, processes, analyzes data, and
takes action if required
• Several non-invasive sensors worn on body• Vital signs data collected and passed (via
gateway) to system database• Database stores, processes, analyzes data, and
takes action if required
QoLT Lab QoLT Lab
increasingincreasing youryour QoLQoL
InternetInternet
Gateway
Monitoring center
Doctor’s office
Remote Monitoring of Vital SignsRemote Monitoring of Vital Signs
QoLT Lab QoLT Lab
increasingincreasing youryour QoLQoLApril 19, 2023 QuBIT Lab's Proprietary 6
Symptom Recognition
Pre-Hospital
Delay in Initiation of Therapy
ED Cath LabCall to Medical System
Increasing Loss of Heart Muscle
Treatment Delayed is Treatment DeniedTreatment Delayed is Treatment Denied
Individual with Chest Pain (CP)
Biggest Challenge
5 min < 30 min < 90 min
Current Practice Standard
30 ± 2.3 h
ECG Sensor Node ImplementationECG Sensor Node Implementation
• Plug-and-Play ECG sensor node for Body Area Network• Connected to PC via USB port
ECG Signal ProcessingECG Signal Processing• ECG preprocessing and feature extraction• LabVIEW’s wavelet toolset used
• ECG preprocessing and feature extraction• LabVIEW’s wavelet toolset used
Baseline Wandering Removal
Wide-band Noise
SuppressionECG Records
Denoising
QRS Complexes Extraction
Fiducial Point Extraction
Preprocessed ECG Data • Wavelet peak and valley detector
• Adaptive Thresholding • Search-back algorithm for possible
missed peaks • Valleys right before and after each
peak (R) determine Q and S points
• Wavelet Detrend • Wavelet Analysis sym5 wavelet
To ECG Beat Classification
Original ECG Signal
Denoised ECG Signal and QRS complexes
marked
Preprocessed ECG Data
• Overall accuracy of 99.51% achieved on MIT-BIH Arrhythmia Database
QRS Duration
ST Segment
ST Interval
PR
PR Interval
Mean R-peak Average RR IntervalMean Power Spectral Density Autocorrelation ValueArea under QRS
One Feature Vector for each Heart Beat
Heart Beat Classification Module Using Heart Beat Classification Module Using Support Vector MachineSupport Vector Machine
Classified Heart Beats
Learning Algorithm-- Support Vector Machine