Artificial intelligence-powered biomedical imaging: Recent achievements and challenges

In recent years, there have been remarkable advancements in artificial intelligence (AI) techniques, particularly in their application to biomedical imaging. This integration has opened up new possibilities for early and improved diagnosis, automation, and interoperability across various medical applications. This review explores the key developments in AI-driven biomedical imaging, examining the techniques and applications that have evolved. We highlight recent enhancements in various areas, such as early-stage diagnostics and explainability. Additionally, we address the challenges and limitations while shedding light on potential research directions to further integrate AI into clinical imaging, thereby enhancing patient-centred care. By synthesizing these key advancements and ongoing challenges, we aim to underscore AI’s potential to transform biomedical imaging practices. • Recent progress in AI-powered biomedical imaging include specialized medical foundation models and vision transformers that have improved segmentation, classification, and multimodal data integration across different modalities, showcasing innovative technological trends that are reshaping the field. • AI-powered solutions offer significant improvements in image analysis, enabling faster and more accurate interpretations, which is critical for timely medical interventions. • Advances in explainable AI are bridging the gap between technical outputs and clinically meaningful interpretations, fostering trust in AI-assisted decision-making. • The review highlights the latest developments in AI for biomedical imaging, emphasizing key challenges and considerations, such as privacy preservation, bias mitigation, and clinical workflow integration.