BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20261101T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260625T210607Z UID:1EB2B7CD-3A81-4370-9682-0AEA298AD1D2 DTSTART;TZID=America/New_York:20260623T140000 DTEND;TZID=America/New_York:20260623T151500 DESCRIPTION:BIOGRAPHY\nDan Istrate is a Professor at the University of Tech nology of Compiègne\, one of the first generalist\nengineering schools in France. He is also a researcher at the Biomechanics and Bioengineering\nL aboratory (BMBI\, UMR CNRS 7338). Since January 2026\, he has served as De puty Director of the\nBMBI Laboratory.\nHis three main research areas are the detection of frailty and distress in older adults\, birth prediction\n in high-risk pregnancies\, and stress detection in ambulatory conditions. He has published 37 articles\nin international journals and presented 95 c ommunications at international conferences. He has\nsupervised 16 PhD stud ents\, 10 postdoctoral researchers\, and 25 master’s students. He has al so\nserved as an expert reviewer for the French National Research Agency\, the French Research Tax\nCredit\, ECOS SUD\, NSERC/CRSNG\, and the AAL Pr ogramme.\n\nABSTRACT\nThe rise of connected medical devices\, the Internet of Medical Things (IoMT)\, and sensor technologies has\nmade Ambient Assi sted Living (AAL) and home monitoring of older adults a critical research area. This work\nfocuses on recognizing Activities of Daily Living (ADL) u sing classical sensors\, such as motion and door-opening\nsensors\, couple d with an original smart sound sensor that recognizes environmental sounds while preserving\npersonal privacy. The system follows a “3N” approac h: no audio recording\, no cloud computing\, and no speech\nrecognition. A hybrid approach based on two layers\, ontology and AI\, is currently bein g developed for ADL\nrecognition.\nFor high-risk pregnancies\, premature b irth can have negative consequences for the newborn. We are\ndeveloping a delivery-date prediction algorithm based on uterine electrical monitoring\ , known as\nelectrohysterography (EHG)\, using multiple surface electrodes . New signal-processing methods and AI-based\nprediction algorithms are be ing proposed.\nStress detection in ambulatory conditions is also a current challenge. Although many devices already exist\,\nincluding smartwatches\ , mobile applications\, and smart wristbands\, stress scales and evaluatio n procedures\nare not yet standardized. Our work focuses on developing met hods based on physiological signals\, including\nheart-rate variability (H RV)\, electrodermal activity (EDA)\, and skin temperature\, to propose a p ersonalized\nstress score and detect chronic stress. We are currently reco rding a real-life database with students during\nwritten and oral examinat ion periods\, coupled with questionnaires\n\nSpeaker(s): Dan Istrate\n\nRo om: 3142\, Bldg: EIT \, 200 University Ave W\, Waterloo\, Ontario\, Canada \, N2L 3G1 LOCATION:Room: 3142\, Bldg: EIT \, 200 University Ave W\, Waterloo\, Ontari o\, Canada\, N2L 3G1 ORGANIZER:amansoor@uwaterloo.ca SEQUENCE:32 SUMMARY:Connected Biomedical Objects for eHealth (IomT) URL;VALUE=URI:https://events.vtools.ieee.org/m/564467 X-ALT-DESC:Description: <br /><p><strong>BIOGRAPHY</strong><br>Dan Istrate is a Professor at the University of Technology of Compi&egrave\;gne\, one of the first generalist&nbsp\;<br>engineering schools in France. He is als o a researcher at the Biomechanics and Bioengineering&nbsp\;<br>Laboratory (BMBI\, UMR CNRS 7338). Since January 2026\, he has served as Deputy Dire ctor of the&nbsp\;<br>BMBI Laboratory.<br>His three main research areas ar e the detection of frailty and distress in older adults\, birth prediction &nbsp\;<br>in high-risk pregnancies\, and stress detection in ambulatory c onditions. He has published 37 articles&nbsp\;<br>in international journal s and presented 95 communications at international conferences. He has&nbs p\;<br>supervised 16 PhD students\, 10 postdoctoral researchers\, and 25 m aster&rsquo\;s students. He has also&nbsp\;<br>served as an expert reviewe r for the French National Research Agency\, the French Research Tax&nbsp\; <br>Credit\, ECOS SUD\, NSERC/CRSNG\, and the AAL Programme.<br><br><stron g>ABSTRACT</strong><br>The rise of connected medical devices\, the Interne t of Medical Things (IoMT)\, and sensor technologies has&nbsp\;<br>made Am bient Assisted Living (AAL) and home monitoring of older adults a critical research area. This work&nbsp\;<br>focuses on recognizing Activities of D aily Living (ADL) using classical sensors\, such as motion and door-openin g&nbsp\;<br>sensors\, coupled with an original smart sound sensor that rec ognizes environmental sounds while preserving&nbsp\;<br>personal privacy. The system follows a &ldquo\;3N&rdquo\; approach: no audio recording\, no cloud computing\, and no speech&nbsp\;<br>recognition. A hybrid approach b ased on two layers\, ontology and AI\, is currently being developed for AD L&nbsp\;<br>recognition.<br>For high-risk pregnancies\, premature birth ca n have negative consequences for the newborn. We are&nbsp\;<br>developing a delivery-date prediction algorithm based on uterine electrical monitorin g\, known as&nbsp\;<br>electrohysterography (EHG)\, using multiple surface electrodes. New signal-processing methods and AI-based&nbsp\;<br>predicti on algorithms are being proposed.<br>Stress detection in ambulatory condit ions is also a current challenge. Although many devices already exist\,&nb sp\;<br>including smartwatches\, mobile applications\, and smart wristband s\, stress scales and evaluation procedures&nbsp\;<br>are not yet standard ized. Our work focuses on developing methods based on physiological signal s\, including&nbsp\;<br>heart-rate variability (HRV)\, electrodermal activ ity (EDA)\, and skin temperature\, to propose a personalized&nbsp\;<br>str ess score and detect chronic stress. We are currently recording a real-lif e database with students during&nbsp\;<br>written and oral examination per iods\, coupled with questionnaires</p> END:VEVENT END:VCALENDAR