BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Talks.cam//talks.cam.ac.uk// X-WR-CALNAME:Talks.cam BEGIN:VEVENT SUMMARY:Cambridge MedAI Seminar - November 2025 - Rui Li and Rehan Zuberi DTSTART:20251126T114500Z DTEND:20251126T130000Z UID:TALK240130@talks.cam.ac.uk CONTACT:Hannah Clayton DESCRIPTION:Join us for the *Cambridge AI in Medicine Seminar Series*\, ho sted by the *Cancer Research UK Cambridge Centre* and the *Department of R adiology at Addenbrooke’s*. This series brings together leading experts to explore cutting-edge AI applications in healthcare—from disease diagn osis to drug discovery. It’s a unique opportunity for researchers\, prac titioners\, and students to stay at the forefront of AI innovations and en gage in discussions shaping the future of AI in healthcare.\n\nThis month ’s seminar will be held on *Wednesday 26 November 2025\, 12-1pm at the J effrey Cheah Biomedical Centre (Main Lecture Theatre)\, University of Camb ridge* and *streamed online via Zoom*. A light lunch from Aromi will be se rved from 11:45. The event will feature the following talks:\n\n*_LUMEN - A deep learning pipeline for analysis of the 3D morphology of the cerebral lenticulostriate arteries from time-of-flight 7T MRI_ - Rui Li\, PhD stud ent\, Department of Clinical Neurosciences\, University of Cambridge*\n\nR ui Li is a PhD student at the Stroke Research Group\, led by Professor Hug h Markus\, in the Department of Clinical Neurosciences\, University of Cam bridge. Her doctoral research focuses on applying machine learning to neur oimage analysis in cerebral small vessel disease research. Specifically\, her work involves developing methods for the segmentation and quantificati on of the morphology of small cerebral perforating arteries from 7T MRI\, and applying machine learning to dementia prediction in cerebral small ves sel disease from multimodal MRI. Prior to her PhD\, she studied informatio n engineering and bioengineering in the Department of Engineering at Cambr idge.\n\n*Abstract*: The lenticulostriate arteries (LSAs) supply critical subcortical brain structures and are affected in cerebral small vessel dis ease (CSVD). Changes in their morphology are linked to cardiovascular risk factors and may indicate early pathology. 7T Time-of-Flight MR angiograph y (TOF-MRA) enables clear LSA visualisation. We aimed to develop a semi-au tomated pipeline for quantifying 3D LSA morphology from 7T TOF-MRA in CSVD patients.\n\n\nWe used data from a local 7T CSVD study to create a pipeli ne\, LUMEN\, comprising two stages: vessel segmentation and LSA quantifica tion. For segmentation\, we fine-tuned a deep learning model\, DS6\, and c ompared it against nnU-Net and a Frangi-filter pipeline\, MSFDF. For quant ification\, centrelines of LSAs within basal ganglia were extracted to com pute branch counts\, length\, tortuosity\, and maximum curvature. This pip eline was applied to 69 subjects\, with results compared to traditional an alysis measuring LSA morphology on 2D coronal maximum intensity projection (MIP) images.\n\n\nFor vessel segmentation\, fine-tuned DS6 achieved the highest test Dice score (0.814±0.029) and sensitivity\, whereas nnU-Net a chieved the best balanced average Hausdorff distance and precision. Visual inspection confirmed that DS6 was most sensitive in detecting LSAs with w eak signals. Across 69 subjects\, the pipeline with DS6 identified 23.5 ± 8.5 LSA branches. Branch length inside the basal ganglia was 26.4 ± 3.5 mm\, and tortuosity was 1.5 ± 0.1. Extracted LSA metrics from 2D MIP anal ysis and our 3D analysis showed fair-to-moderate correlations. Outliers hi ghlighted the added value of 3D analysis.\n\n\nThis open-source deep-learn ing-based pipeline offers a validated tool quantifying 3D LSA morphology i n CSVD patients from 7T-TOF-MRA for clinical research.\n\n\n*_Multimodal L earning to Predict Progression in Barrett’s Oesophagus_ - Rehan Zuberi\, PhD student\, Cancer Research UK Cambridge Institute\, University of Camb ridge*\n\nRehan Zuberi is a PhD researcher at the Cancer Research UK Cambr idge Institute in the Markowetz Lab. His work focuses on developing machin e learning architectures and applying them to cancer research\, with an em phasis on building clinically relevant multimodal models for early detecti on of disease.\n\n*Abstract*: Oesophageal cancer is one of the deadliest c ancers\, with most patients not surviving beyond five years. Early detecti on is therefore critical\, but current surveillance methods often miss sub tle progression signals. This talk will present a multimodal deep learning framework that integrates whole slide histopathology features with genomi c copy number variation (CNV) data to predict progression in Barrett’s o esophagus. We use weakly supervised multiple instance learning for image r epresentation and an intermediate fusion architecture to combine modalitie s. I will discuss dataset composition\, fusion strategies\, and early perf ormance benchmarks\, as well as the unique signal captured beyond unimodal baselines. I will also outline future work exploring additional modalitie s such as clinical and longitudinal data. This approach aims to improve ri sk stratification and enable earlier intervention for patients at high ris k of progression.\n\nThis is a hybrid event so you can also join via Zoom: \n\nhttps://zoom.us/j/99050467573?pwd=UE5OdFdTSFdZeUtIcU1DbXpmdlNGZz09\n\n Meeting ID: 990 5046 7573 and Passcode: 617729\n\nWe look forward to your participation! If you are interested in getting involved and presenting yo ur work\, please email Ines Machado at im549@cam.ac.uk\n\nFor more informa tion about this seminar series\, see: https://www.integratedcancermedicine .org/research/cambridge-medai-seminar-series/ LOCATION:Jeffrey Cheah Biomedical Centre (Main Lecture Theatre)\, Universi ty of Cambridge END:VEVENT END:VCALENDAR