Towards trustworthy digital patient monitoring: deep learning approaches for analyzing multimedia data from the actual clinical practice

Title

Towards trustworthy digital patient monitoring: deep learning approaches for analyzing multimedia data from the actual clinical practice

Speaker

Lucia Migliorelli - Università Politecnica delle Marche

Abstract

In this seminar, we explore the emerging field of digital patient monitoring, emphasizing the importance of trustworthiness in healthcare applications. We delve into the use of deep learning techniques for the analysis of multimedia data directly derived from real-world clinical settings. Our focus is on how these advanced methods may enhance the efficiency of patient-monitoring systems, thereby supporting clinicians in making informed decisions. We address the challenges associated with processing multimedia data, including images and audio recordings. Additionally, we discuss the ethical considerations and the need for implementing bias mitigation measures from the early stages of software design to distribute trustworthy technologies.

Bio

Lucia Migliorelli is a post-doctoral researcher at the Università Politecnica delle Marche (UNIVPM) and co-founder of AIDAPT srl. She specializes in the development of deep-learning algorithms for analyzing multimedia data to support clinical decision-making processes. She is a co-author of publications in international journals and conferences, and a lecturer in computer vision and deep learning for the master's degree in computer and automation engineering at the UNIVPM. In addition to her work, she dedicates herself to scientific dissemination through workshops and conferences for middle and high school students, promoting accessibility to STEM disciplines and stimulating interest in science from a young age.

When

Monday January 15th, 16:00

Where

Room 704, DIMA, Via Dodecaneso 35