ASGE Annual Postgraduate Course: Clinical Challenges and Real-Time Decision Making in GI Endoscopy (On-demand) | May 2023-AI in predictive modeling: Complex EHR Data

Video Transcription

Video Summary

In this video, TAPOCHE Roy, senior manager at Kaiser Permanente, discusses the role of computer vision and endoscopy in analyzing big data sets and predictive modeling. He begins by explaining the different perspectives on electronic health records (EHR) from clinicians, IT professionals, and data scientists, highlighting the need for insights at both individual and population levels. Roy emphasizes the complexity of healthcare data, mentioning a framework by Dr. Thomas Kannapalli and team for understanding complexity in healthcare. He then delves into the life cycle of predictive modeling, explaining the steps involved in developing and testing models for classification and regression problems. Roy provides examples of classification and regression models in gastroenterology and mentions clustering as a means of segmentation in unsupervised learning. He also briefly touches on reinforcement learning and its potential applications in healthcare. Roy concludes by discussing the current opportunities for improvement in machine learning, such as fairness, bias, transparency, explainability, and trustability. He introduces an EHR solution that includes a cohort manager, lineage mapper, and visualization capabilities for both population and individual views. Roy provides references for further reading and invites questions from the audience.

Asset Subtitle

Taposh Roy

Keywords

computer vision

predictive modeling

healthcare data complexity

unsupervised learning

machine learning improvement