BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251016T135143Z
LOCATION:Riverside East
DTSTART;TZID=America/Chicago:20251014T173000
DTEND;TZID=America/Chicago:20251014T183000
UID:HFESAM_ASPIRE 2025_sess131_POST364@linklings.com
SUMMARY:Towards Generalizable Drowsiness Monitoring with Physiological Sen
 sors: A Preliminary Study
DESCRIPTION:Jiyao Wang (The Hong Kong University of Science and Technology
  (Guangzhou), The Hong Kong University of Science and Technology); Suzan A
 yas (University of Toronto); Jiahao Zhang (The Hong Kong University of Sci
 ence and Technology (Guangzhou)); Xiao Wen (The Hong Kong University of Sc
 ience and Technology); Dengbo He (The Hong Kong University of Science and 
 Technology (Guangzhou)); and Birsen Donmez (University of Toronto)\n\nAccu
 rately detecting drowsiness is vital to driving safety. Among different me
 asures, physiological-signal-based drowsiness monitoring can be more priva
 cy-preserving than a camera-based approach. However, conflicts exist regar
 ding how physiological metrics are associated with different drowsiness la
 bels across datasets, which might reduce the generalizability of data-driv
 en models trained with multiple datasets. Thus, we analyzed key features f
 rom electrocardiograms (ECG), electrodermal activity (EDA), and respirator
 y (RESP) signals across four datasets, where different drowsiness inducers
  (such as fatigue and low arousal) and assessment methods (subjective vs. 
 objective) were used. Binary logistic regression models were built to iden
 tify the physiological metrics that are associated with drowsiness. Findin
 gs indicate that distinct drowsiness inducers can lead to different physio
 logical responses, and objective assessments were more sensitive than subj
 ective ones in detecting drowsiness. Further, decreased heart rate stabili
 ty, respiratory amplitude, and tonic EDA are robustly associated with incr
 eased drowsiness. These results enhance the understanding of drowsiness de
 tection and can inform future generalizable monitoring designs.\n\n
END:VEVENT
END:VCALENDAR