Title: THREADS: A Molecular-driven Foundation Model for Oncologic Pathology

Abstract: Foundation models are reshaping computational pathology by enabling transfer learning, where models pre-trained on vast datasets can be adapted for downstream diagnostic, prognostic, and therapeutic response tasks. Despite these advances, foundation models are still limited in their ability to encode the entire gigapixel whole-slide images without additional training and often lack complementary multimodal data. In this talk, I will introduce THREADS, a slide-level foundation model capable of generating universal representations of whole-slide images of any size. THREADS was pretrained using a multimodal learning approach on a diverse cohort of 47,171 hematoxylin and eosin (H&E)-stained tissue sections, paired with corresponding genomic and transcriptomic profiles—the largest such paired dataset to be used for foundation model development to date. This unique training paradigm enables THREADS to capture the tissue’s underlying molecular composition, yielding powerful representations applicable to a wide array of downstream tasks. In extensive benchmarking across 54 oncology tasks, including clinical subtyping, grading, mutation prediction, immunohistochemistry status determination, treatment response prediction and survival prediction THREADS outperformed all baselines while demonstrating remarkable generalizability and label efficiency. It is particularly well-suited for predicting rare events, further emphasizing its clinical utility.

Bio: Anurag Vaidya is a fourth-year PhD student in the Health Sciences Technology program at Harvard and MIT, where he is supervised by Dr. Faisal Mahmood from Harvard Medical School. Anurag completed his undergraduate studies in biomedical engineering and computer science at Bucknell University. His doctoral work focuses on developing tools that integrate multimodal clinical data in supervised and unsupervised ways to improve cancer diagnosis, prognosis, and biomarker discovery.

Paper Link: [2501.16652] Molecular-driven Foundation Model for Oncologic PathologyLink opens in a new windowLink opens in a new windowLink opens in a new window

How to attend: Either turn up to the event on the day, or if you want to attend online then please contact Adam Shephard (adam.shephard@warwick.ac.uk) for more details.

Title: Towards ecologically sustainable AI for pathology

Abstract: Artificial Intelligence (AI) is already widespread in medical research and will expectedly revolutionize clinical practice and health care. AI methods can be employed to analyze large medical data collections, to diagnose disease or to perform prognosis of disease progression. This is particularly true for pathology, a medical diagnostic field mainly relying on image data Pathology might be one of the first areas of medicine substantially transformed by AI. At the same time, the growing energy consumption and CO2 emissions associated with both development and large-scale application of AI models become a concern. AI promises benefits for the patient, but comes at the cost of accelerating climate change. In this talk, I will discuss the environmental consequences of using AI in pathology. I will also touch upon potential approaches to consider and mitigate this towards a more environmentally friendly “green” AI.

Bio: Professor Peter Boor received his medical and scientific training at the Medical Schools of Bratislava in Slovakia and Aachen Germany. He is the chair of Translational Nephropathology and senior consultant pathologist at the Institute of Pathology at the RWTH Aachen University. He also leads the Electron Microscopy Facility and Digital Pathology and is the RWTH Lecturer. He is a member of several national and international societies of pathology, renal pathology, and nephrology and received several prestigious awards and prizes for academic, clinical, and teaching excellence. His research group, the LaBooratory of Nephropathology, focuses on diagnostic biomarkers with a particular focus on imaging, digital pathology, and AI, as well as in vivo animal modeling, and understanding of pathological processes in CKD, fibrosis, and (micro)vasculature. He founded and is leading the German National Autopsy Registry (NAREG) and Research Network (NATON) and other research consortia. His scientific work encompasses more than 340 original papers, reviews and editorials, and several book chapters.

Paper Link: Ecologically sustainable benchmarking of AI models for histopathology | npj Digital MedicineLink opens in a new windowLink opens in a new windowLink opens in a new window

How to attend: Either turn up to the event on the day, or if you want to attend online then please contact Adam Shephard (adam.shephard@warwick.ac.uk) for more details.

Title: A foundation model for joint segmentation, detection and recognition of biomedical objects across nine modalities

Abstract: Biomedical image analysis is fundamental for biomedical discovery. Holistic image analysis comprises interdependent subtasks such as segmentation, detection and recognition, which are tackled separately by traditional approaches. Here, we propose BiomedParse, a biomedical foundation model that can jointly conduct segmentation, detection and recognition across nine imaging modalities. This joint learning improves the accuracy for individual tasks and enables new applications such as segmenting all relevant objects in an image through a textual description. To train BiomedParse, we created a large dataset comprising over 6 million triples of image, segmentation mask and textual description by leveraging natural language labels or descriptions accompanying existing datasets. We showed that BiomedParse outperformed existing methods on image segmentation across nine imaging modalities, with larger improvement on objects with irregular shapes. We further showed that BiomedParse can simultaneously segment and label all objects in an image. In summary, BiomedParse is an all-in-one tool for biomedical image analysis on all major image modalities, paving the path for efficient and accurate image-based biomedical discovery.

Bio: Theodore Zhao is a Senior Applied Scientist at Microsoft Health and Life Sciences Research, working on multimodal biomedical imaging models as well as biomedical natural language processing. Theodore earned his PhD in Applied Mathematics degree from University of Washington, where his research applied machine learning, stochastic modeling and optimization to applications in finance and healthcare. His research interests focus on machine learning, self-supervised learning, multimodal models, and mathematical modeling.

Paper Link: A foundation model for joint segmentation, detection and recognition of biomedical objects across nine modalities | Nature MethodsLink opens in a new windowLink opens in a new window

How to attend: Either turn up to the event on the day, or if you want to attend online then please contact Adam Shephard (adam.shephard@warwick.ac.uk) for more details.

Title: AI Agents in Oncology

Abstract: AI agents expand upon traditional AI systems by autonomously executing multi-step tasks in cancer research and oncology. These systems can interact with software, plan iteratively, and suggest therapeutic strategies or optimize drug design with minimal human input. AI agents show promise in automating intellectual tasks in cancer research and supporting clinical decision-making in oncology. Significant challenges remain regarding regulatory frameworks, interpretability, and ethical considerations for healthcare implementation. This talk provides a primer on AI agents in cancer research and oncology, examining their capabilities, applications, and constraints.

Bio: Professor Jakob Kather holds dual appointments in medicine and computer science at the Technical University (TU) Dresden, Germany, serves as a senior physician in medical oncology at the University Hospital Dresden and holds an additional affiliation with the National Center for Tumor Diseases (NCT) in Heidelberg. His research is focused on applying artificial intelligence in precision oncology. Prof. Kather’s research team at TU Dresden is using deep learning techniques to analyze a spectrum of clinical data, including histopathology, radiology images, textual records, and multimodal datasets. Guided by the belief that medical and tech expertise needs to be combined, medical researchers in his team learn computer programming and data analysis, while computer scientists are immersed in cancer biology and oncology. Prof. Kather chairs the “Working group on Artificial Intelligence” at the German Society of Hematology and Oncology (DGHO) and is a member of the pathology task force of the American Association for Cancer Research (AACR). His work is supported by numerous European and national grants, which enable the team to develop new deep learning methods for medical data analysis techniques and to apply them in precision oncology.nk opens in a new windowLink opens in a new windowLink opens in a new window

How to attend: Either turn up to the event on the day, or if you want to attend online then please contact Adam Shephard (adam.shephard@warwick.ac.uk) for more details.