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:20260202T201229Z
LOCATION:Second Floor Atrium
DTSTART;TZID=America/Chicago:20251118T080000
DTEND;TZID=America/Chicago:20251118T170000
UID:submissions.supercomputing.org_SC25_sess537_drs104@linklings.com
SUMMARY:OAAgent: A Multimodal LLM Agent Clinical Assistant for Precision O
 steoarthritis Care
DESCRIPTION:Pegah Ahadian (Kent State University)\n\nOsteoarthritis (OA) i
 s a chronic condition which affects over 300 million people globally and i
 s a leading cause of disability, yet predictive models often remain monomo
 dal, static, and opaque to clinicians. This dissertation develops OAAgent,
  a multimodal large language model (LLM) clinical assistant that integrate
 s medical images (X-ray, MRI), longitudinal clinical variables, and physic
 ian notes for personalized, interpretable OA care and prediction of progre
 ssion. OAAgent employs a fusion transformer for multimodal integration, a 
 temporal retrieval system C-TRAG with explicit cross-visit semantics to re
 trieve clinically similar cases (similar past trajectories), and reinforce
 ment learning for dynamic decision-making through a Chain-of-Thought reaso
 ning layer and Extract-and-Abstract clinical note summarization to ensure 
 transparent, patient-specific recommendations. The dissertation addresses 
 five critical gaps at the intersection of OA AI research:\n\n1. Joint inte
 gration of heterogeneous modalities\n2. Longitudinal temporal reasoning\n3
 . Clinically interpretable decision support\n4. Personalized treatment rec
 ommendations\n5. Inclusion of underutilized narrative notes\n\nOAAgent’s a
 rchitecture is designed for extensibility through the Model Context Protoc
 ol (MCP), enabling it to interoperate with other domain-specific models, m
 ultimodal pipelines, and external reasoning agents. This creates a bridge 
 between the LLM core and diverse analytical components, enhancing adaptabi
 lity to new modalities and clinical contexts.\n\nDeveloped in collaboratio
 n with the Cleveland Clinic and validated on the OAI dataset and FNIH coho
 rt and MIMIC datasets, OAAgent demonstrates improved accuracy, temporal ca
 libration, and interpretability. Anchored in a Trustworthy AI framework, t
 his work advances agentic multimodal AI for healthcare, offering a scalabl
 e, ethical, and interoperable pathway toward equitable, explainable clinic
 al decision support across chronic diseases.\n\nTag: Research & ACM SRC Po
 sters\n\nRecording: Not Livestreamed, Not Recorded\n\nRegistration Categor
 y: Technical Program Reg Pass\n\nSession Chairs: Kento Sato (RIKEN Center 
 for Computational Science (R-CCS)); Chris Schlipalius (Pawsey Supercomputi
 ng Research Centre; Commonwealth Scientific and Industrial Research Organi
 sation (CSIRO), Australia); and Anja Gerbes (Georg-August-Universität Gött
 ingen)\n\n
END:VEVENT
END:VCALENDAR