Key result
Machine learning models achieve an AUC of 0.80 for predicting VTE in cancer patients.
Why the study?
Recent studies have suggested that machine learning could accurately predict venous thromboembolism in cancer patients, but its overall predictive performance required evaluation.
Do machine learning models accurately predict venous thromboembolism in patients with cancer?
Systematic Review (n=12,249)
Do machine learning models accurately predict venous thromboembolism in patients with cancer?
Effect estimate: AUC 0.80
Machine learning models demonstrate good accuracy (test set AUC 0.80) for predicting venous thromboembolism in cancer patients, though external validation is needed.
May aid VTE risk stratification in cancer; extends ML evidence but requires prospective validation before adoption.
PURPOSE: Recent studies have suggested that machine learning (ML) could be used to predict venous thromboembolism (VTE) in cancer patients with high accuracy. METHODS: We aimed to evaluate the performance of ML in predicting VTE events in patients with cancer. PubMed, Web of Science, and EMBASE to identify studies were searched. RESULTS: Seven studies involving 12,249 patients with cancer were included. The combined results of the different ML models demonstrated good accuracy in the prediction of VTE. In the training set, the global pooled sensitivity was 0.87, the global pooled specificity was 0.87, and the AUC was 0.91, and in the test set 0.65, 0.84, and 0.80, respectively. CONCLUSION: The prediction ML models showed good performance to predict VTE. External validation to determine the result's reproducibility is necessary.
No takes yet. Share an insight, caveat, or question.
Franco‐Moreno et al. (2023) conducted a systematic review in Cancer (n=12,249). Machine learning models was evaluated on Prediction of venous thromboembolism (AUC in test set) (AUC 0.80). Machine learning models demonstrated good accuracy in predicting venous thromboembolism in cancer patients, achieving a pooled AUC of 0.80 in the test set.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: