BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
DTSTART:20250309T030000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20251102T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251005T031850Z
UID:09919AEB-1E04-415E-810F-6C2CF1B29165
DTSTART;TZID=America/New_York:20251001T150000
DTEND;TZID=America/New_York:20251001T153000
DESCRIPTION:Medical image segmentation is crucial for precise anatomical de
 lineation in diagnostic and therapeutic procedures. Despite significant ad
 vancements in medical image segmentation models\, achieving both high accu
 racy and fairness remains a challenging and underexplored area\, as improv
 ements in one metric often lead to reductions in the other. This study add
 resses the challenge of enhancing both accuracy and fairness in segmentati
 on by mitigating demographic biases through supervised curriculum learning
  and progressive loss. We employ a manually annotated dataset from the Ost
 eoarthritis Initiative (OAI)\, including both hip and knee radiographs. By
  applying various curriculum learning strategies and distinct progressive 
 loss functions that shift focus from easier to more challenging examples\,
  we aim to improve the models' accuracy and fairness. By considering demog
 raphic factors such as race and gender\, we evaluate and mitigate biases i
 n segmentation outcomes\, leading to enhanced segmentation accuracy. Our f
 indings contribute to the advancement of medical image analysis and the pr
 omotion of fair AI models for healthcare applications.\n\nCo-sponsored by:
  XDI Lab (www.xdilab.com)\n\nVirtual: https://events.vtools.ieee.org/m/504
 152
LOCATION:Virtual: https://events.vtools.ieee.org/m/504152
ORGANIZER:hmoradi@ncat.edu
SEQUENCE:8
SUMMARY:Enhancing Segmentation Fairness Through Curriculum Learning and Pro
 gressive Loss
URL;VALUE=URI:https://events.vtools.ieee.org/m/504152
X-ALT-DESC:Description: <br /><p><span style="font-family: 'Segoe UI'\; fon
 t-size: 11.25pt\;">Medical image segmentation is crucial for precise anato
 mical delineation in diagnostic and therapeutic procedures. Despite signif
 icant advancements in medical image segmentation models\, achieving both h
 igh accuracy and fairness remains a challenging and underexplored area\, a
 s improvements in one metric often lead to reductions in the other. This s
 tudy addresses the challenge of enhancing both accuracy and fairness in se
 gmentation by mitigating demographic biases through supervised curriculum 
 learning and progressive loss. We employ a manually annotated dataset from
  the Osteoarthritis Initiative (OAI)\, including both hip and knee radiogr
 aphs. By applying various curriculum learning strategies and distinct prog
 ressive loss functions that shift focus from easier to more challenging ex
 amples\, we aim to improve the models' accuracy and fairness. By consideri
 ng demographic factors such as race and gender\, we evaluate and mitigate 
 biases in segmentation outcomes\, leading to enhanced segmentation accurac
 y. Our findings contribute to the advancement of medical image analysis an
 d the promotion of fair AI models for healthcare applications.</span></p>
END:VEVENT
END:VCALENDAR