AI applications in lumbar and lumbosacral pedicle screw placement: a systematic review of limited evidence and future directions.
Authors
Affiliations (6)
Affiliations (6)
- Program in Translational Medicine, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Samut Prakan, Thailand.
- Faculty of Medicine Ramathibodi Hospital, Chakri Naruebodindra Medical Institute, Mahidol University, Samut Prakan, Thailand.
- Airport of Thailand Public Company Limited, Bangkok, Thailand.
- Bachelor of Engineering (Biomedical Engineering), Faculty of Engineering, Royal Melbourne Institute of Technology University, Melbourne, Australia.
- Bachelor of Biological Sciences (Biomedical Sciences Module), Faculty of Science, Mahidol University International College, Nakhon Pathom, Thailand.
- Neurosurgery Division, Surgery Department, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand. [email protected].
Abstract
Artificial intelligence (AI) is a general term that refers to the use of a computer to simulate intelligent behavior with minimal human intervention. Currently, AI can be applied to various spine surgery approaches. This review aims to provide a clearer picture of AI’s applicability for the perioperative period and enhance outcomes for pedicle screw fixation (PS). The PRISMA guideline was applied, which identified 14 studies regarding AI applications in PS. We categorized the AI application to PS into segmentation, object detection, image registration, and other categories, such as improved quality and converted images. Then, an analysis and discussion of the current trends and applications of various AI models in PS methods was performed. The effects of AI performance included a reduction in the time required for operations and planning, automatic identification of screws and anatomical landmarks, reduced image errors, and reduced radiation exposure. However, the lack of training data and less data diversity remain the limitations of model development, as both factors impact model generalization and robustness. This data extraction might reveal research gaps, providing researchers with ideas for future studies regarding AI and PS integration for better medical care outcomes. The online version contains supplementary material available at 10.1007/s10143-026-04192-2.