MICCAI Workshop on Personalized Incremental Learning in Medicine

Held in conjunction with the 27th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) – October 10, 2024, Marrakesh (Morocco) 

Web site: https://miccai-pilm.github.io

About

Machine learning models for personalized medicine are designed to customize diagnosis and treatment based on individual patient data. These models often deal with complex, sparse, and diverse data and typically require a large, joint dataset from multiple patients to create a universal model that can be adjusted for individual patients. However, collecting such comprehensive datasets is challenging due to data governance issues, and historical data may not be reliable due to changes in medical equipment and diagnostic methods over time.

The workshop on Personalized Incremental Learning in Medicine (PILM) aims to bridge the gap between incremental learning research and its application to personalized medicine, allowing machine learning models to learn from new data gradually while retaining previously acquired knowledge. This is particularly useful in personalized medicine, as it enables models to be trained on data from just a few or even a single patient, enhancing privacy and allowing for ongoing updates to the models as new patient data becomes available.

Topics

Potential topics include, but are not limited to:

– Novel algorithms for incremental and continual learning that are suitable for medical applications.

– Methods to prevent catastrophic forgetting in the context of patient-specific machine learning models.

– Strategies for one-shot or few-shot learning in medical diagnosis and treatment personalization.

– Techniques for handling domain shifts within a patient’s data over time or across different medical devices.

– Approaches to integrate incremental learning with transfer learning in medicine.

– Evaluation metrics and methodologies for assessing the performance of incremental learning systems in personalized medicine.

– Ethical considerations and data privacy solutions in the development of incremental learning models for healthcare.

– Case studies and practical applications of incremental learning in medical imaging, patient monitoring, and other areas of personalized medicine.

– Discussions on the limitations of current datasets and proposals for new data collection efforts that support incremental learning research in medicine.

– Interdisciplinary research that combines insights from clinical practice, medical imaging, and machine learning to advance personalized medicine.

Important dates

Paper submission deadline: June 30, 2024

Notification to authors: July 15, 2024

Camera-ready deadline: July 30, 2024

Proceedings

Accepted papers will be published in Springer LNCS in a separate proceedings book.

Organizers

Simone Palazzo (University of Catania, Italy)

Giovanni Bellitto (University of Catania, Italy)

Nancy Zlatintsi (National Technical University of Athens, Greece)

Panagiotis Filntisis (National Technical University of Athens, Greece)

Cecilia S. Lee (University of Washington)

Aaron Y. Lee (University of Washington)

We hope to see you in Marrakesh!