Development and validation of a machine learning-based risk prediction model for cancer-related fatigue in ovarian cancer patients
By
Ru Feng
Zexuan Fan
Yuanyuan Pang
Qifan Ding
Qian Yue
Siqi Wei
June 4, 2026
Clinical Scorecard: Creation and assessment of a machine learning-driven model for predicting cancer-related fatigue in patients with ovarian cancer
At a Glance
Category Detail
Condition Cancer-related fatigue (CRF) in ovarian cancer Key Mechanisms Interaction of biological, psychological, and social factors Target Population Ovarian cancer patients Care Setting Tertiary hospitals
Key Highlights
CRF prevalence in ovarian cancer patients is 39.6%. Support Vector Machine (SVM) model achieved AUC of 0.884. Top predictive features include serum calcium level and anxiety-depression status. Machine learning models provide personalized predictions and strong adaptability. Early identification of high-risk patients enables targeted interventions. Guideline-Based Recommendations
Diagnosis
Utilize machine learning algorithms for CRF risk prediction. Management
Implement targeted interventions based on CRF risk stratification. Monitoring & Follow-up
Regular assessment of CRF symptoms and associated factors. Risks
Consider biological, psychological, and social factors influencing CRF. Patient & Prescribing Data
Ovarian cancer patients aged 18 and older.
CRF is associated with treatment side effects and impacts quality of life.
Clinical Best Practices
Employ machine learning models for comprehensive CRF assessment. Focus on psychosocial support to mitigate CRF impacts. Related Resources & Content
