Advances in Weak Supervision for Medical Image Analysis

Dear Coleagues,

Recent advances in machine and deep learning have revealed significant progress in multiple medical image analysis tasks, such as registration, reconstruction, classification, etc. Most approaches require the collection of a large amount of annotated data and their performance notably decreases when only a small amount of data are available or when the data prove to be heterogenous, unbalanced, or incomplete. Moreover, the process of annotation is tedious and complex, even for experts (medical doctors for instance), and this limits the accumulation of good-quality data.

To address these issues, a promising strategy consists of developing machine learning models that have a good performance in solving image analysis problems, but with a partial annotation of data, leading to the weak supervision of the learning phase.

The purpose of this Special Issue is to collect and diffuse recent advances in research proposing novel solutions to face these challenges. These can concern the development of weak supervised machine learning models for medical image analysis as well as other related subjects, such as the improvement of data quality, efficient strategies for annotating data, and research about the explainability of deep learning architectures. This latter topic is of high importance for creating weak supervised models, since it permits us to understand the capacities of fully supervised models according to datasets (the importance of features, classification errors, incorrect data annotation, etc.) and potentially reduce their need for supervision.  

Prof. Dr. Antoine Vacavant
Prof. Dr. Anna Fabijańska
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Journal of Imaging is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• medical image analysis and processing (segmentation, classification, detection, registration, …)
• machine/deep learning
• weak supervision
• data annotation, quality of data
• explainability of deep learning