Related Experiment Video
Updated: Jun 18, 2025
09:16
Multi-Modal Signals for Analyzing Pain Responses to Thermal and Electrical Stimuli
Published on: April 5, 2019
10.6K
Artificial Intelligence and Machine Learning in Cancer Pain: A Systematic Review
Vivian Salama1, Brandon Godinich2, Yimin Geng3
1Department of Radiation Oncology (V.S., B.G., L.H.V., L.M., K.A.W., M.A.N., R.H., A.S.R.M., C.D.F., A.C.M), The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
Journal of Pain and Symptom Management
|August 3, 2024
Summary
Artificial intelligence and machine learning (AI/ML) show promise in predicting and managing cancer pain. However, further research is needed to improve model quality and clinical validation for reliable application.
Area of Science:
- Oncology
- Medical Informatics
- Data Science
Background:
- Cancer pain is a complex symptom affecting most patients.
- Artificial intelligence/machine learning (AI/ML) offers potential solutions for cancer pain management.
- This review explores AI/ML applications in predicting cancer pain outcomes.
Purpose of the Study:
- To systematically review AI/ML applications in predicting cancer pain outcomes.
- To evaluate AI/ML models for cancer pain management.
- To assess the quality and performance of AI/ML models in cancer pain research.
Main Methods:
- Comprehensive database search (Ovid MEDLINE, EMBASE, Web of Science) up to September 7, 2023.
- Inclusion of studies on AI/ML in cancer pain prediction and management.
- Quality assessment using PROBAST risk-of-bias and TRIPOD guidelines.
Main Results:
- 44 studies (2006-2023) utilized AI/ML for cancer pain.
- AI/ML models showed varying performance (median AUC 0.77), with Random Forest models performing best (median AUC 0.81).
- High risk-of-bias (77.3%) and limited external validation (14%) were noted, alongside poor reporting of model calibration (5%).
Conclusions:
- AI/ML tools hold significant potential for advancing cancer pain classification, risk stratification, and management.
- Improving model quality, calibration, and conducting rigorous external validation are crucial.
- Ensuring practical and reliable clinical application of AI/ML in cancer pain management requires further research.
