Evolving Paradigms in Medical Image Analysis: From CNNs to Foundation Models

Dear Colleagues,

Medical image analysis remains one of the most impactful and technically demanding domains in artificial intelligence. The unique physical properties, acquisition protocols, and dimensional characteristics of medical imaging modalities require specialized computational approaches.

The volumetric structure of modalities such as CT and MRI, as well as the temporal dynamics of ultrasound and fluoroscopy, create opportunities for advanced modeling strategies, including Transformer-based architectures which have demonstrated strong potential in modeling long-range spatial and temporal dependencies. In parallel, foundation models pretrained on large-scale imaging or multimodal datasets are increasingly being adapted for healthcare applications to mitigate the challenges of limited annotated medical data.

Despite these advancements, convolutional neural networks (CNNs) continue to play a central role in this field. The relative advantages, limitations, and practical trade-offs between traditional deep learning architectures and modern large-scale models remain insufficiently characterized across many clinical scenarios.

This Special Issue aims to provide a rigorous and comparative platform for studies employing CNNs, Transformers, hybrid architectures, and foundation models in medical imaging. We particularly welcome contributions that offer methodological innovation, clinical validation, benchmarking, interpretability analysis, or deployment considerations.

We invite researchers working at the intersection of artificial intelligence and medical imaging to submit original research articles and comprehensive reviews.

In this Special Issue, original research articles and review papers are welcome.

Research areas may include (but are not limited to):

• Comparative studies of CNNs vs. transformers in medical imaging
• Adaptation and fine-tuning of foundation models for healthcare
• Self-supervised and transfer learning in low-data medical settings
• Multimodal imaging and multimodal learning approaches
• 3D and 4D medical image modeling
• Vision-language models in radiology and pathology
• Model interpretability and explainability in clinical imaging
• Generalization, domain adaptation, and federated learning
• Computational efficiency and deployment considerations
• Benchmark datasets and reproducibility studies

We look forward to receiving your contributions.

Dr. Elham Mahmoudi
Dr. Allison Scarbrough
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 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-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 imaging
• convolutional neural networks
• transformers
• foundation models
• deep learning
• self-supervised learning
• multimodal learning
• computational imaging