Speaker "Shubham Patil" Details Back



Explainability and interpretability in Medical AI


The use of AI-enabled technologies in healthcare has significantly improved decision-making by healthcare staff, including pathologists, radiologists, and surgeons, leading to better patient outcomes. However, medical AI faces unique challenges, particularly in achieving explainability and interpretability. As AI algorithms become more complex, understanding how they make decisions is critical. This is especially true in medical AI, where model safety and trustworthiness rely on interpretability and explainability. The presentation will discuss the challenges associated with achieving interpretability and explainability in medical AI, including the complexity of data, the opaque nature of some AI algorithms, and the need to balance accuracy with interpretability. Additionally, it will examine the potential benefits of interpretable and explainable AI in medical decision-making, such as improved patient outcomes, reduced costs, and increased transparency. Furthermore, we will explore the latest methods, techniques, and research used to overcome these challenges.

Who is this presentation for?
The presentation will be geared toward the technical crowd interested in developing AI solutions in healthcare.

Prerequisite knowledge:
Basics of Machine Learning and deep learning

What you'll learn?
Deep dive into the medical AI landscape, explainability/interpretability in medical settings using AI, trends, methods, and techniques.


Shubham is a Staff Deep Learning Engineer with Stryker's Robotics and AI division. He concentrates on enhancing patient outcomes, promoting access to quality healthcare, and advancing the state of healthcare via developing AI-enabled technologies and devices. Prior to his current role, Shubham led the deployment of edge devices for Computer Vision-based AI algorithms that recognized human actions in high-pressure healthcare settings while working at DawnLight. He holds a master's degree in Robotics from Carnegie Mellon University's School of Computer Science.