Related Experiment Video
Updated: Jan 8, 2026
06:24
A Spine Robotic-Assisted Navigation System for Pedicle Screw Placement
Published on: May 11, 2020
9.3K
Artificial intelligence in spine surgery: a scoping review.
Anis Choucha1, Morgane Evin2, Matteo de Simone3
1Aix Marseille Univ, APM, UH Timone, Department of Neurosurgery, Marseille, France; Laboratory of Biomechanics and Application, UMRT24, Gustave Eiffel University, Aix Marseille University, Marseille, France.
Neuro-Chirurgie
|December 15, 2025
Summary
Artificial intelligence (AI) in spinal surgery is developing, but limited external validation and data access hinder adoption. Increased collaboration and transparency are crucial for integrating AI into clinical practice.
Area of Science:
- Spinal Surgery
- Artificial Intelligence
- Medical Technology
Background:
- Artificial intelligence (AI) integration in spinal surgery offers potential for evidence-based and personalized patient care.
- Current applications of AI in daily surgical practice are still in early development stages.
- This scoping review maps AI applications, identifies current research frontiers, and highlights literature gaps in spinal surgery.
Purpose of the Study:
- To conduct a scoping review of artificial intelligence applications in spinal surgery.
- To map the current landscape and frontiers of AI in spinal surgery research.
- To identify existing gaps in the literature regarding AI in spinal surgery.
Main Methods:
- A comprehensive scoping review following PRISMA guidelines was performed.
- Searches were conducted in PubMed and Cochrane databases up to January 2024.
- Studies detailing AI models or validated AI applications in spinal surgery were included, extracting data on objectives, outcomes, models, validation, disease types, institutions, and journals.
Main Results:
- The United States led contributions (32%), followed by China (18%) and Europe (15%).
- Degenerative (24%) and oncological (19%) conditions were the most studied pathologies, with deep learning methods dominating.
- While 76% of studies reported validation, rigorous external validation and resource sharing (data/code access) were limited.
Conclusions:
- Widespread adoption of AI in spinal surgery is impeded by a lack of rigorous external validation and restricted access to models and datasets.
- Enhanced interdisciplinary collaboration and transparency in AI model development are necessary.
- Making validated AI models publicly accessible is essential to bridge the gap for broader clinical integration.
