How can we automate the analysis of medical images? This review provides a critical assessment of semi-automated and automated methods for segmenting anatomical medical images. Image segmentation automates or facilitates the delineation of anatomical structures and regions of interest, a crucial step in many medical-imaging applications. The authors present terminology and important issues in image segmentation. They then review current approaches, emphasizing the advantages and disadvantages of these methods for medical imaging. Techniques reviewed include region-based methods, edge detection, clustering, and model-based segmentation. This study also covers the advantages and disadvantages of current approaches. The authors conclude with a discussion on the future of image segmentation methods in biomedical research. The review provides a valuable resource for researchers and practitioners seeking to improve the accuracy and efficiency of medical image analysis. With its comprehensive overview of current techniques and its focus on biomedical research, this paper is set to be a seminal work in its field.
As a review in the Annual Review of Biomedical Engineering, this article is appropriately contextualized. It provides a comprehensive overview of medical image segmentation, a core topic in biomedical engineering. This assessment of current methodologies will help shape future innovation.