SpineFM: leveraging foundation models for automatic spine X-ray segmentation

Conference item

This paper introduces SpineFM, a novel pipeline that achieves state-of-the-art performance in the automatic segmentation and identification of vertebral bodies in cervical and lumbar spine radiographs. SpineFM leverages the regular geometry of the spine, employing a novel inductive process to sequentially infer the location of each vertebra along the spinal column. Vertebrae are segmented using Medical-SAM-Adaptor, a robust foundation model that diverges from commonly used CNN-based models. We achieved outstanding results on two publicly available spine X-Ray datasets, with successful identification of 97.8% and 99.6% of annotated vertebrae, respectively. Of which, our segmentation reached an average Dice of 0.942 and 0.921, surpassing previous state-of-the-art methods.

Authors: Simons, SJ; Papiez, BW

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: IEEE
• Host title: IEEE ISBI 2025 Symposium Proceedings
• Pages: 1-4
• Publication date: 2025-05-12
• Acceptance date: 2024-12-17
• Event title: 2025 IEEE International Symposium on Biomedical Imaging
• Event location: Houston, TX, USA
• Event website: https://biomedicalimaging.org/2025/
• Event start date: 2025-04-14
• Event end date: 2025-04-17
• DOI: 10.1109/ISBI60581.2025.10980854
• EISSN: 1945-8452
• ISSN: 1945-7928
• EISBN: 9798331520526
• ISBN: 9798331520526
• Language: English
• Keywords: vertebra segmentation; radiography; geometry; image segmentation; foundation models; biological system modeling; spine; pipelines; X-ray imaging; biomedical imaging; spinal X-ray imaging