Related Experiment Video
Updated: Feb 22, 2026
08:41
A MRI-Based Toolbox for Neurosurgical Planning in Nonhuman Primates
Published on: July 17, 2020
5.4K
Natural and Artificial Intelligence in Neurosurgery: A Systematic Review
Joeky T Senders1,2, Omar Arnaout2,3, Aditya V Karhade2
1Department of Neurosurgery, University Medical Center, Utrecht, the Netherlands.
Neurosurgery
|September 26, 2017
Summary
Machine learning (ML) models show potential in neurosurgery, often outperforming clinical experts in accuracy. Collaboration between humans and ML is key to overcoming implementation challenges.
Area of Science:
- Artificial Intelligence
- Neurosurgery
Background:
- Machine learning (ML) enables algorithms to learn from data without explicit programming.
- This study reviews ML applications in neurosurgery compared to clinical expertise.
Purpose of the Study:
- To summarize neurosurgical applications of ML compared to clinical expertise.
- To assess ML's performance against neurosurgical clinical expertise.
Main Methods:
- A systematic literature search of PubMed and Embase was conducted up to August 2016.
- Studies comparing ML algorithms with clinical experts in neurosurgery were reviewed.
Main Results:
- Twenty-three studies utilized ML for diagnosis, planning, or prediction in neurosurgery.
- ML models showed median improvements of 13% in accuracy and 0.14 in AUC compared to experts.
- ML outperformed experts in 58% of outcome measures; 36% showed no difference; experts outperformed ML in 6%.
Conclusions:
- ML models can enhance clinical decision-making in neurosurgery.
- Hurdles in ML model creation, validation, and deployment must be addressed.
- A human-and-machine collaborative approach is crucial for successful clinical integration.
