4EU+ online seminars on "Artificial Intelligence Techniques, Applications, and Social Issues"
The importance of explainable AI models in the context of biomedical data
11 February 2021 16:00 – 17:30 CET Link to online seminars: Zoom
Abstract
The advent of deep learning (DL) has led to a significant step forward in the performance of applications related to computer vision, speech recognition, natural language processing, etc. This result was mainly achieved by largely increasing the model complexity and amount of data, while compromising the understanding about what the model does. However, other scientific fields must trade off between performance and possibility of interpreting (or understanding) the AI model’s predictions. Biomedical applications are among those in need. In this seminar, we will talk about the motivations behind such trade off, review the current state-of-the-art methodologies for interpreting predictions for DL models and discuss about innovative technique of Explainable AI in the context of electrocardiography.
Short bio
Dr. Massimo Walter Rivolta is an assistant professor and member of the Biomedical image and Signal Processing Group at the University of Milan, Italy. He received the M.Sc. degree in Biomedical Engineering from the Politecnico di Milano, Italy, and the Ph.D degree in Computer Science from the University of Milan, Italy. His research interests include signal processing, optimization techniques, feature extraction, computational intelligence and computerized simulations, with interest in biomedical applications, specifically for the creation of automatic tool for processing of electrocardiographic signals, heart-rate variability series and accelerometer signals.

4EU+ online seminars on "Artificial Intelligence Techniques, Applications, and Social Issues"

The importance of explainable AI models in the context of biomedical data

11 February 2021
16:00 – 17:30 CET

Link to online seminars: Zoom

Abstract

The advent of deep learning (DL) has led to a significant step forward in the performance of applications related to computer vision, speech recognition, natural language processing, etc. This result was mainly achieved by largely increasing the model complexity and amount of data, while compromising the understanding about what the model does. However, other scientific fields must trade off between performance and possibility of interpreting (or understanding) the AI model’s predictions. Biomedical applications are among those in need. In this seminar, we will talk about the motivations behind such trade off, review the current state-of-the-art methodologies for interpreting predictions for DL models and discuss about innovative technique of Explainable AI in the context of electrocardiography.

Short bio

Dr. Massimo Walter Rivolta is an assistant professor and member of the Biomedical image and Signal Processing Group at the University of Milan, Italy. He received the M.Sc. degree in Biomedical Engineering from the Politecnico di Milano, Italy, and the Ph.D degree in Computer Science from the University of Milan, Italy.

His research interests include signal processing, optimization techniques, feature extraction, computational intelligence and computerized simulations, with interest in biomedical applications, specifically for the creation of automatic tool for processing of electrocardiographic signals, heart-rate variability series and accelerometer signals.