Machine Learning for Predicting Post-Operative Outcomes in Meningiomas
Author Information
Author(s): Abualnaja Siraj Y., Morris James S., Rashid Hamza, Cook William H., Helmy Adel E.
Primary Institution: University of Cambridge
Hypothesis
Can machine learning algorithms accurately predict post-operative outcomes in meningioma patients?
Conclusion
Machine learning algorithms have the potential to improve the accuracy of prognoses for meningioma patients.
Supporting Evidence
- Machine learning models incorporating both clinical and radiomic data significantly outperformed traditional methods.
- Pooled estimates for the AUCs achieved by different ML algorithms ranged from 0.74–0.81.
- Ensemble classifiers showed promise for future clinical application.
Takeaway
This study shows that computers can help doctors predict how well patients will do after surgery for brain tumors called meningiomas.
Methodology
A systematic review and meta-analysis of 32 studies using random-effects meta-analysis to evaluate machine learning algorithms.
Potential Biases
Potential bias in predictive accuracy towards female patients due to the higher prevalence of meningiomas in females.
Limitations
Many studies did not compare machine learning models with traditional methods, and there was significant heterogeneity in the data.
Participant Demographics
Mean age of patients was 56.03 years, with a male-to-female ratio of 37:63.
Statistical Information
P-Value
p=0.04
Confidence Interval
0.74–0.81
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website