Machine Learning Models for Predicting Pain, Fatigue, Depression, Anxiety, and Malnutrition in Cancer Patients: A Systematic Review and Meta-Analysis.
Maidenamu Reheman, Yunhuan Li, Yang Chen et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis evaluated machine learning models for predicting pain, fatigue, depression, anxiety, and malnutrition in cancer patients. A total of 34 studies were included, revealing pooled AUCs of 0.76 for pain, 0.82 for fatigue, 0.76 for depression, 0.78 for anxiety, and 0.86 for malnutrition, indicating acceptable predictive performance. The study highlights the need for further validation.
Key findings
- Pooled AUC for predicting pain was 0.76 (95% CI: 0.69-0.83).
- Pooled AUC for predicting fatigue was 0.82 (95% CI: 0.76-0.88).
- Pooled AUC for predicting malnutrition was 0.86 (95% CI: 0.80-0.91).
- Moderate certainty of evidence was noted across all outcomes.
Why it matters
Understanding the predictive capabilities of machine learning models for cancer-related symptoms can enhance symptom management and improve quality of life for patients. Clinicians may leverage these models to identify at-risk patients and tailor interventions accordingly.