SBME Research Seminar - Dr. Hannah Carter
Immune Checkpoint Blockade (ICB) has revolutionized cancer treatment, however mechanisms determining patient response remain poorly understood. We used machine learning to predict ICB response from germline and somatic biomarkers and studied feature usage by the learned model to uncover putative mechanisms driving superior outcomes. Patients with higher T follicular helper infiltrates were robust to defects in the class-I Major Histocompatibility Complex (MHC-I). Further investigation uncovered different ICB responses in MHC-I versus MHC-II neoantigen reliant tumors across patients. Despite similar response rates, MHC-II reliant responses were associated with significantly longer durable clinical benefit (Discovery: Median OS=63.6 vs. 34.5 months P=0.0074; Validation: Median OS=37.5 vs. 33.1 months, P=0.040). Characteristics of the tumor immune microenvironment reflected MHC neoantigen reliance, and analysis of immune checkpoints revealed LAG3 as a potential target in MHC-II but not MHC-I reliant responses. This study highlights the value of interpretable machine learning models in elucidating the biological basis of therapy responses.
Events
Calendar
- This event has passed.
Research Seminar: Using the past to predict the future: Building and deploying AI Models in the Perioperative Period – Dr. Ira Hofer
February 8, 2024 @ 11:00 am - 12:00 pm PST
Research Seminar: Using the past to predict the future: Building and deploying AI Models in the Perioperative Period – Dr. Ira Hofer
Talk Summary:
In this talk participants will gain perspective on the history of medicine, why augmented intelligence (AI) is key to the future of managing health, the current state of perioperative AI models, and the changes necessary to make the future a reality.
Dr. Ira Hofer | Department of Anethesiology and Medicine; Icahn School of Medicine at Mount Sinai
Dr. Ira Hofer Biography:
Ira S. Hofer, MD is a practicing anesthesiologist and clinical informaticist at The Icahn School of Medicine at Mount Sinai. Dr. Hofer has been working in the field of big data analytics and electronic health record (EHR) data for over 15 years. He has had a variety of roles at multiple hospital systems building and deploying AI models and overseeing analytics for system wide and departmental programs. He currently holds joint appointments in Anesthesiology, Medicine and the Charles Bronfman Institute for Personalized Medicine and is the inaugural director of the Rapid Randomized Trial Unit in the CBIM which is focused on deploying AI models into clinical care using study designs to measure efficacy.
Dr. Hofer’s research interests involve leveraging the data from the EHR to understand, quantify and ultimately mitigate risk in the perioperative period. His work includes some of the first papers to apply machine learning techniques for perioperative outcome prediction and has been featured on the cover of both Anesthesiology and Anesthesia&Analgesia. Ira has a K01 from the National Heart Lung and Blood institute, and funding from the NIA funded IMPACT Collaboratory for a pragmatic trial to reduce postoperative delirium using an AI driven decision support model. He also serves as an associate editor for Anesthesia & Analgesia. His current research focus is on creating featurization techniques to incorporate a wide range of EHR data into machine learning models in order to improve discrimination and calibration as well as establishing multi-center perioperative collaboratives to better share raw and processed EHR data.
In addition to his academic accomplishments, Dr. Hofer is the Founder and President of Extrico Health based on technology he developed. The Extrico Software Platform has been successfully deployed at major hospitals around the nation and helps providers turn the raw data from their electronic health record into actionable insights to improve patient care, increase operational efficiencies and optimize the revenue cycle.
Join Virtually:
Zoom Meeting ID: 91264 571213
Password: 571213
Location:
Life Sciences Centre
LSC 1003 (LT3)