Upcoming Talks

Opportunistic Screening for Pancreatic Cancer – Multimodal AI Fusion of CT Imaging and Radiology Reports

Pancreatic ductal adenocarcinoma (PDAC) is among the most lethal cancers, with most cases diagnosed at advanced stages and limited treatment options. Early detection is critical but remains elusive due to subtle imaging signatures and a lack of reliable biomarkers. In this talk, I will present our work on opportunistic screening using routinely acquired CT imaging and corresponding radiology reports from patients prior to PDAC diagnosis. We developed a multimodal survival model that integrates quantitative radiomics features with contextual information from clinical narratives using deep language embeddings. By fusing text and imaging data, our model achieved improved prognostic accuracy and risk stratification across internal and external cohorts compared to single-modality baselines. These findings highlight the potential of multimodal AI to identify high-risk individuals earlier, guide clinical decision-making, and advance precision medicine in pancreatic cancer care.

Dr. David Le is a Postdoctoral Fellow at the Mayo Clinic in Phoenix, Arizona. His research bridges medical imaging, artificial intelligence (AI), and clinical translation, with a focus on developing digital biomarkers and survival models for early detection and prognosis. His work spans pancreatic cancer imaging and ophthalmic applications, integrating deep learning, radiomics, and vision–language models. By combining quantitative imaging with AI-driven risk modeling, he aims to improve patient stratification, enable earlier interventions, and advance precision medicine.

Coming soon

Please add your questions to the speaker either to this google form or directly under the YouTube video