Machine Learning and Deep Learning in Medical Imaging

Dear Colleagues,

In today’s healthcare landscape, the importance of computer-aided diagnosis is rapidly growing, offering clear benefits to both medical professionals and patients. The automation of processes traditionally managed by humans is becoming increasingly prominent, particularly in medical image analysis. By harnessing networks capable of multilayer pattern analyses—collectively referred to as artificial intelligence (AI)—and large datasets containing input data, these systems can provide results with minimal error bias. AI algorithms play a pivotal role in medical imaging, especially in tasks such as pattern detection, recognition, and result description. Applications based on deep learning and machine learning enhance key decision-making steps, including image segmentation, classification, and transforming results into natural language.

Building on the success of previous topics, Artificial Intelligence in Medical Imaging and Image Processing and AI in Medical Imaging and Image Processing, which attracted numerous submissions and published articles, this continuation seeks to delve deeper into advancements in this dynamic field. This topic welcomes contributions focusing on image processing and recognition powered by deep learning and machine learning, not only in radiology but also across other medical disciplines. Additionally, papers presenting AI-based solutions designed to standardize diagnostic processes and significantly reduce the time required for pathology detection and description are highly encouraged.

By showcasing diverse AI techniques and their applications across various fields of medicine, this topic aims to promote knowledge exchange and enhance the understanding of both technical aspects and practical implementations of AI in contemporary medical imaging.

Aims:

1. Advance Knowledge in AI-Powered Medical Imaging:
• Facilitate a deeper understanding of how machine learning (ML) and deep learning (DL) transform medical imaging by enhancing accuracy, speed, and reliability.
2. Encourage Multidisciplinary Collaboration:
• Bridge the gap between AI researchers, clinicians, radiologists, and healthcare professionals to foster innovative solutions.
3. Highlight Practical Implementations:
• Showcase real-world applications of AI in diagnostics, pathology detection, and patient care to emphasize the tangible impact of these technologies.
4. Promote Innovation in Diagnostic Standardization:
• Explore AI-based frameworks for standardized diagnostic processes, reducing subjectivity and error bias.
5. Drive Research on Emerging Techniques:
• Investigate advanced deep learning models, such as convolutional neural networks (CNNs), transformers, and generative models, in medical image analysis.

Scope:

1. Medical Image Processing Techniques:
• Contributions focusing on advanced methods for image segmentation, classification, and enhancement;
• Integration of AI tools for real-time processing of complex medical imaging modalities such as CT, MRI, ultrasound, and X-rays.
2. Multidisciplinary Applications:
• AI applications across various medical fields beyond radiology, such as dermatology, ophthalmology, pathology, and cardiology.
3. Novel AI Frameworks and Algorithms:
• Development and application of innovative AI architectures tailored to specific imaging challenges, such as low-resolution imaging or noisy data.
4. Natural Language Integration:
• Research on transforming imaging results into natural language reports, bridging communication gaps between AI systems and healthcare providers.
5. Ethical and Regulatory Considerations:
• Discussions on the ethical use, data privacy, bias mitigation, and regulatory frameworks for AI in medical imaging.
6. Standardization and Scalability:
• Papers that propose scalable AI-driven solutions to streamline diagnostic workflows and ensure consistent outcomes across healthcare systems.
7. AI for Pathology Detection and Monitoring:
• AI techniques for the early detection, progression tracking, and treatment monitoring of diseases, including cancer, cardiovascular disorders, and neurological conditions.
8. Real-Time Diagnostics:
• Research on AI-driven solutions enabling real-time, point-of-care diagnostics with minimal human intervention.

Dr. Rafał Obuchowicz
Prof. Dr. Michał Strzelecki
Prof. Dr. Adam Piorkowski
Dr. Karolina Nurzynska
Topic Editors