Abstract
Accurate segmentation of both the pituitary gland and adenomas from magnetic resonance imaging (MRI) is essential for diagnosis and treatment of pituitary adenomas. This systematic review evaluates automatic segmentation methods for improving the accuracy and efficiency of MRI-based segmentation of pituitary adenomas and the gland itself. We analysed 34 studies that employed automatic and semi-automatic segmentation methods out of 353 reviewed studies. We extracted and synthesized data on segmentation techniques and performance metrics (such as Dice overlap scores). The majority of reviewed studies utilized deep learning approaches, with U-Net-based models being the most prevalent. Automatic methods yielded Dice scores of 0%-89% for pituitary gland and 4%-96% for adenoma segmentation. Semi-automatic methods reported 80%-92% for pituitary gland and 75%-88% for adenoma segmentation. Most studies did not report important metrics such as MR field strength, age and adenoma size (macro/micro/giant) or even adenoma type and human subject numbers. Automated segmentation techniques such as U-Net-based models show promise, especially for adenoma segmentation, but further improvements are needed to achieve consistently good performance in small structures like the normal pituitary gland. Future progress will require methodological innovation and larger, more diverse datasets to enhance clinical applicability. Systematic Review Registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD42023407127, PROSPERO CRD42023407127.