Skip to main content Skip to secondary navigation
Main content start

AIMI-IBIIS Seminar: Fusion of Multi-Modal Data Stream for Clinical Event Prediction – Simulating Physician’s Workflow - Imon Banerjee, PhD

Event Details:

Wednesday, April 21, 2021
12:00pm - 1:00pm PDT

AIMI YouTube Channel

Open to all

Link to join or set a reminder: bit.ly/2P9X5Zk

Abstract:
Advancements in machine learning and deep learning techniques carry the potential to make significant contributions to healthcare, particularly in fields of diagnosis, prognosis, and treatment decisions. However, most of current state-of-the-art machine learning and deep learning models for healthcare applications consider only a single input data stream without data informing clinical context. The trend of ignoring clinical contextual information is particularly prominent when dealing with the diagnosis and prognosis tasks where the imaging data is accessible. Yet in practice, pertinent and accurate non-imaging data based on the clinical history and laboratory data enable physicians to interpret imaging findings in the appropriate clinical context, leading to a higher diagnostic accuracy, informative clinical decision making, and improved patient outcomes. To achieve a similar goal using machine learning and increase the physician trust, clinical diagnosis and prognosis models must also achieve the capability to process contextual clinical data from electronic health records (EHR) in addition to pixel or other sensor data. This talk will present multiple fusion machine learning models on the imaging data with boosted performance by integrating the clinical context. In addition to imaging, I will also present a smart flexible sensor patch with on-chip AI capability that can be used in homecare to generate advance alert of cardiovascular abnormality by combining physiological signal data with patient demographic and comorbidity information.

About:
Imon Banerjee is an Assistant Professor in Emory University with joint affiliation with Biomedical Informatics and Radiology, and an active member of Winship Cancer Institute. She is also working as a Co-director of Medical and Health Informatics Core (MHIC) at Emory University. Her background is in computer science (CS) and she received the prestigious Marie Curie EU fellowship during her doctoral study. She completed her postdoctoral training from the Biomedical Data Science Department at Stanford University. Currently, she is jointly running HITILab which hosts more that 30 CS graduate, undergrad, radiology residency and international CS and EE intern students. She is also leading multiple federal and nonfederal grants related to predictive modeling and biomedical image analysis. Her research is focused on unstructured medical data analysis (mainly clinical notes and images) and integration of multisource medical data from varying hospital systems for building predictive models to benefit cancer diagnosis and treatment. She is currently leading multiple innovative multi-institutional research projects which involves both academic (Duke, Stanford, Harvard, Intermountain, SUNY) and industrial (Philips, GE healthcare) partners.

Related Topics

Explore More Events