BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Cambridge AI in Medicine Seminar - April 2026 - Lingjia Wang and S
 huaiyu Yuan
DTSTART:20260428T104500Z
DTEND:20260428T120000Z
UID:TALK246436@talks.cam.ac.uk
CONTACT:Hannah Clayton
DESCRIPTION:Sign up on Eventbrite: https://medai-april2026.eventbrite.co.u
 k\n\nJoin us for the *Cambridge AI in Medicine Seminar Series*\, hosted by
  the *Cancer Research UK Cambridge Centre* and the *Department of Radiolog
 y at Addenbrooke’s*. This series brings together leading experts to expl
 ore cutting-edge AI applications in healthcare – from disease diagnosis 
 to drug discovery. It’s a unique opportunity for researchers\, practitio
 ners\, and students to stay at the forefront of AI innovations and engage 
 in discussions shaping the future of AI in healthcare.\n\nThis month’s s
 eminar will be held on *Tuesday 28 April 2026\, 12-1pm at the Jeffrey Chea
 h Biomedical Centre (Main Lecture Theatre)\, University of Cambridge* and 
 *streamed online via Zoom*. A light lunch from Aromi will be served from 1
 1:45. The event will feature the following talks:\n\n*_AI-derived Whole-li
 ver PDFF for Quantitative Assessment of Hepatic Steatosis_ - Lingjia Wang\
 , PhD Candidate\, Department of Radiology\, University of Cambridge*\n\nLi
 ngjia is a second-year PhD candidate in the Department of Radiology at the
  University of Cambridge\, working on AI-based disease prediction and oppo
 rtunistic screening in CT and MRI. Lingjia's research particularly focuses
  on translating automated imaging biomarkers into PACS-integrated clinical
  workflows.\n\n*Abstract*: Hepatic steatosis exhibits spatially heterogene
 ous fat distribution that may be incompletely captured by routine region-o
 f-interest (ROI)–based proton density fat fraction (PDFF) measurements. 
 This retrospective study evaluated an AI-derived whole-liver PDFF framewor
 k in 556 abdominal liver MRI examinations (2023–2025). Automated whole-l
 iver and Couinaud segment masks were generated using TotalSegmentator on 3
 D Dixon images and applied to PDFF maps\, excluding voxels <0% or >55%. Ag
 reement between whole-liver and radiologist-reported ROI PDFF was assessed
  using Bland–Altman analysis\, and intrahepatic fat heterogeneity was an
 alysed across Couinaud segments. Whole-liver PDFF showed good agreement wi
 th ROI PDFF (mean bias −0.35%\; limits of agreement −5.48% to +4.78%).
  Segmental analysis demonstrated greater variability and lower mean PDFF i
 n the left lobe compared with the right lobe. AI-derived whole-liver PDFF 
 provides a standardised assessment of hepatic fat and captures spatial het
 erogeneity beyond single-ROI measurements.\n\n*_TMoE: Task-Conditioned Mix
 ture-of-Experts with Task-Query Heads for Medical Image Classification_ - 
 Shuaiyu Yuan\, PhD Candidate\, Department of Radiology\, University of Cam
 bridge*\n\n*Abstract*: Medical image classification can cover heterogeneou
 s datasets from various modalities\, dimensionalities\, anatomical focuses
  and di agnostic tasks. While training separate models is costly\, perform
 ing a combined supervised learning with a fully shared backbone often suff
 ers from negative transfer and limited feature extraction capacity. We pro
  pose TMoE\, a Task-Conditioned Mixture-of-Experts with Task-Query Heads f
 or multi-task medical image classification. TMoE augments a vi sion Transf
 ormer backbone by replacing feed-forward networks (FFN) in selected Transf
 ormer blocks with sparse MoE layers. A Top-k router\, conditioned on a lea
 rned task embedding\, dispatches tokens to subsets of experts\, enabling t
 ask-dependent capacity allocation. We also intro duce task-query heads\, w
 here a per-task learned query performs cross attention over backbone token
 s to produce a task-specific representation\, followed by a lightweight ta
 sk-specific classifier.\n\nThis is a hybrid event so you can also join via
  Zoom: https://zoom.us/j/99050467573?pwd=UE5OdFdTSFdZeUtIcU1DbXpmdlNGZz09\
 n\nMeeting ID: 990 5046 7573 and Passcode: 617729\n\nWe look forward to yo
 ur participation! If you are interested in getting involved and presenting
  your work\, please email Ines Machado at im549@cam.ac.uk\n\nFor more info
 rmation about this seminar series\, see: https://www.integratedcancermedic
 ine.org/research/cambridge-medai-seminar-series/
LOCATION:Jeffrey Cheah Biomedical Centre (Main Lecture Theatre)\, Universi
 ty of Cambridge
END:VEVENT
END:VCALENDAR