Workshop on explainability of machine/deep learning models for medical applications

July 11th, 2024 (2-6 pm)

Organized by the INFORM project

“Interpretability of Deep Neural Networks for Radiomics”

Neo Christopher Chung1, Mathieu Hatt2 and Panagiotis Papadimitroulas3

1 Faculty of Mathematics, Informatics and Mechanics of the University of Warsaw, Poland
2 LaTIM, INSERM UMR 1101, Univ Brest, Brest, France
3 BIOEMTECH, Athens, Greece,

Registration: free

Two types of submissions (format short abstract 3000 char. max) are welcome:

  1. Short abstract describing a study or a project focusing on the development, the evaluation and/or the use of explainability methods or interpretability tools, in the context of medical imaging/radiotherapy, health data, clinical applications.
  2. A use case to present and discuss during the round table in the second part of the workshop to develop an appropriate explainability/interpretability approach dedicated to the use case. The description should cover the application, available data, objective/task, machine/deep learning algorithms used and models developed.

Submissions and/or registrations requests (name, position, affiliation) to be sent to until June 17, 2024


2:00 - 2:30 pm Introductory talk: “Explainability, why it matters”
2:30 - 4:10 pm Talks of submitted abstracts (20’ each, 15’ for presentation, 5’ for discussion)
4:10 - 4:30 pm Coffee break
4:30 - 6:00 pm Round table/open discussion (submitted use cases).
Short presentation of use cases (10min), followed by discussion.

Location of the workshop will be near the ICCR 2024 congress venue.
Bibliothèque Universitaire Lyon 1
Campus de la Doua, 20 avenue Gaston Berger, 69100 Villeurbanne, Lyon France