Trustworthy AI for Medical Image Analysis

Dear Colleagues,

Artificial intelligence, and deep learning in particular, has now reached expert-level performance in many medical imaging tasks, including lesion detection, segmentation, classification, and disease grading across radiology, dermatology, histopathology, ophthalmology, and beyond. Yet the translation of these algorithms into routine clinical practice remains limited. A central obstacle is trust: many models operate as black boxes; produce poorly calibrated confidence estimates; degrade under distribution shifts between scanners, institutions, and patient populations; may encode demographic biases; and are sensitive to image artefacts and adversarial perturbations. Clinicians, patients, and regulators increasingly expect medical AI to be transparent, robust, fair, privacy-preserving, and accountable, not merely accurate.

This Special Issue aims to collect novel algorithmic contributions making medical image analysis systems trustworthy by design. We invite the submission of original research articles, comprehensive reviews, and extended versions of high-quality conference papers (with substantial new material) on topics including, but not limited to, the following:

• Explainable and interpretable models, including saliency and concept-based explanations, prototype learning, and inherently interpretable architectures;
• Uncertainty quantification, confidence calibration, and selective prediction for safer clinical decision support;
• Robustness and generalization under domain shift, including domain adaptation, test-time adaptation, and adversarial robustness;
• Fairness auditing and bias mitigation across demographic and acquisition subgroups;
• Privacy-preserving and federated learning for multi-institutional medical imaging;
• Data-centric trustworthiness: preprocessing, quality control, artefact removal, label-noise handling, and synthetic data generation;
• Reliability of foundation and vision–language models in medical imaging;
• Lightweight and efficient models for reliable deployment at the point of care and on edge devices;
• Evaluation protocols, benchmarks, reproducibility, and human-in-the-loop validation supporting clinical and regulatory readiness.

Submissions should emphasize algorithmic novelty together with rigorous and reproducible evaluation. Authors are strongly encouraged to use public datasets where possible and to release their code, so that the collection can serve as a practical reference for researchers and practitioners working towards clinically dependable AI.

Prof. Dr. Alaa Eleyan
Guest Editor

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 250 words) can be sent to the Editorial Office for assessment.

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-anonymized peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Algorithms 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.