BEGIN:VCALENDAR VERSION:2.0 PRODID:Linklings LLC BEGIN:VTIMEZONE TZID:Europe/Stockholm X-LIC-LOCATION:Europe/Stockholm BEGIN:DAYLIGHT TZOFFSETFROM:+0100 TZOFFSETTO:+0200 TZNAME:CEST DTSTART:19700308T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:+0200 TZOFFSETTO:+0100 TZNAME:CET DTSTART:19701101T020000 RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260724T151410Z LOCATION:Bldg. 8 - B 101 DTSTART;TZID=Europe/Stockholm:20260629T133000 DTEND;TZID=Europe/Stockholm:20260629T153000 UID:submissions.pasc-conference.org_PASC26_sess110@linklings.com SUMMARY:MS1G - Advancing Medical AI: Challenges for Developing AI-Driven I n-Silico Clinical Trials for Accelerating Translational Medicine DESCRIPTION:Access the recording\n\nOrganizer(s): John Garcia-Henao (Balgr ist University Hospital), and Carlos Barrios Hernandez (Universidad Indust rial de Santander, LIG/INRIA - CITI Laboratory)\n\nArtificial intelligence (AI) has achieved major success in medicine, particularly in diagnostic i maging, pathology classification, and clinical report generation, accelera ting research translation and improving care. However, most deployed syste ms remain task-specific, lack biomedical reasoning, and generalize poorly across data modalities and clinical settings. The Advancing Medical AI min isymposium explores how emerging approaches, especially large multimodal m odels (LMMs), can enable AI-driven in-silico clinical trials (ISCTs) that better connect research innovation with clinical application. Recent LMMs integrate medical images, text, and structured data to support diagnosis, segmentation, and reporting, enabling the simulation of biological and cli nical processes and advancing virtual patient modeling. Key challenges rem ain in explainability, computational efficiency, privacy protection, and i ntegration with hospital infrastructure, highlighting the need for transpa rent data governance and verifiable systems. In parallel, ISCTs are gainin g momentum as computer-based experiments that model disease progression an d therapy response in virtual patient cohorts. Built on digital twins-dyna mic computational models continuously updated with clinical data, ISCTs pr omise lower costs, faster development, and improved safety. Despite their potential, barriers such as data heterogeneity, limited interpretability, validation gaps, regulatory constraints, and infrastructure demands persis t.\n\nDigital Twins Meet the HPC Continuum: Distributed Systems Challenges for Scalable and Privacy-Aware In-Silico Medicine\n\nThe convergence of A I and computational modeling is redefining scientific workflows, moving fr om centralized HPC platforms toward a distributed, heterogeneous computing continuum. In healthcare, this evolution is exemplified by medical digita l twins, which integrate multimodal patient data, simulati...\n\n\nFrédéri c Le Mouël (University of Lyon, INSA Lyon; CITI Laboratory)\n------------- --------\nCarbon-Aware Compression Evaluation for Sustainable Medical Imag e Classification\n\nDeep learning models for medical imaging often require substantial computational resources, resulting in high energy consumption and carbon emissions that limit deployment in resource-constrained clinic al environments. We propose a carbon-aware evaluation framework for assess ing deep learning compre...\n\n\nCarlos Barrios Hernandez (Universidad Ind ustrial de Santander, LIG/INRIA - CITI Laboratory)\n---------------------\ nA Hospital-Integrated Digital Twin Ecosystem for Translational Musculoske letal AI\n\nArtificial intelligence is rapidly transforming musculoskeleta l (MSK) medicine through advances in multimodal imaging, segmentation foun dation models, and clinical data integration. However, translating these s ystems from research into clinical practice remains limited by fragmented infrastructures,...\n\n\nJohn Garcia-Henao (Balgrist University Hospital)\ n---------------------\nHPC Infrastructure for Multimodal Medical AI: Less ons from Bridges-2 and Neocortex\n\nLarge multimodal models (LMMs) are rap idly transforming medical AI by enabling integrated analysis across imagin g, clinical text, structured health records, and biomedical data. These mo dels are reshaping how researchers approach diagnostics, clinical decision support, and biomedical discovery by co...\n\n\nPaola Buitrago (Pittsburg h Supercomputing Center)\n\nDomain: Engineering, Life Sciences, Computatio nal Methods and Applied Mathematics\n\nSession Chairs: John Anderson Garci a Henao (University of Bern, ARTORG Center for Biomedical Engineering Rese arch) and Carlos Barrios Hernandez (Universidad Industrial de Santander, L IG/INRIA - CITI Laboratory) END:VEVENT END:VCALENDAR