Overview

Machine learning (ML) systems are achieving remarkable performances at the cost of increased complexity. Deep neural networks, in particular, appear as black box machine and their behaviour can be sometimes unpredictable. Furthermore, more complex models are less interpretable, which may cause distrust. As these systems are pervasively being introduced to critical domains, such as medical image computing and computer assisted intervention (MICCAI), it becomes imperative to develop methodologies for explaining model predictions.

Such methodologies would help physicians to decide whether they should follow/trust a prediction or not, and might help to identify failure cases. Additionally, it could facilitate the deployment of such systems, from a legal perspective. Ultimately, interpretability is closely related with AI safety in healthcare.

However, there is very limited work regarding interpretability of ML systems among the MICCAI research. Besides increasing trust and acceptance by physicians, interpretability of ML systems can be helpful during method development. For instance, by inspecting if the model is learning aspects coherent with domain knowledge, or by studying failures. Also, it may help revealing biases in the training data, or identifying the most relevant data (e.g., specific MRI sequences in multi-sequence acquisitions). This is critical since the rise of chronic conditions has led to a continuous growth in usage of medical imaging, while at the same time reimbursements have been declining. while at the same time reimbursements have been declining. Hence, interpretability can help to improve productivity of image acquisition protocols by highlighting learned features and their relationships to disease patterns.

The Workshop on Interpretability of Machine Intelligence in Medical Image Computing (iMIMIC) at MICCAI 2022 aims at introducing the challenges & opportunities related to the topic of interpretability of ML systems in the context of MICCAI.

Scope

Interpretability can be defined as an explanation of the machine learning system. It can be broadly defined as global, or local. The former explains the model and how it learned, while the latter is concerned with explaining individual predictions. Visualization is often useful for assisting the process of model interpretation. The model’s uncertainty may be seen as a proxy for interpreting it, by identifying difficult instances. Still, although we can find some approaches for tackling machine learning interpretability, there is a lack of formal and clear definition and taxonomy, as well as general approaches, as well as regulatory frameworks. Additionally, interpretability results often rely on comparing explanations with domain knowledge. Hence, there is need for defining objective, quantitative, and systematic evaluation methodologies.

This workshop aims at fostering the discussions, presentation of ideas to tackle the many challenges and identifying opportunities related to the topic of interpretability of ML systems in the context of MICCAI. Therefore, the main purposes of this workshop are:

1. To introduce the challenges/opportunities related to the topic of interpretability of machine learning systems in the context of MICCAI. While there have been workshops of interpretability of machine learning systems in general machine learning and A.I. conferences (NeurIPS, ICML), to the best of our knowledge, iMIMIC is the only workshop dedicated to the medical imaging application domain.
2. To understand the state of the art of this field. This will be achieved through the submitted manuscripts and the invited keynote speakers.
3. To join researchers in this field, and to discuss the issues related to it and future work.
4. To understand the implications of (or lack of) interpretability of machine learning systems in the MICCAI field.

Covered topics include but are not limited to:

• Definition of interpretability in context of medical image analysis.
• Visualization techniques useful for model interpretation in medical image analysis.
• Local explanations for model interpretability in medical image analysis.
• Interpretability methods to make use of multimodal data.
• Causal interpretability.
• Methods to improve transparency of machine learning models commonly used in medical image analysis.
• Textual explanations of model decisions in medical image analysis.
• Uncertainty quantification in context of model interpretability.
• Quantification and measurement of interpretability.
• Legal and regulatory aspects of model interpretability in medicine.

Preliminary Program

The program of the workshop includes keynote presentations of experts working in the field of interpretability of machine learning. A selection of submitted manuscripts will be chosen for long oral presentations (9 minutes + 3 minutes Q&A) and short oral presentations (6 minutes + 3 minutes Q&A) alongside the keynotes. Finally, we will have a group discussion which leaves room for a brainstorming on the most pressing issues in interpretability of machine intelligence in the context of MICCAI.

Preliminary program:

SGT time - September 22nd

Best paper award