A machine learning (ML) algorithm was developed to predict the risk of bleeding among patients with venous thromboembolism (VTE) taking anticoagulation, according to results of a study published in the British Journal of Haematology.

The ML algorithm was generated using 55 baseline variable predictors using data from 49,587 patients in the Registro Informatizado de Engermedad TromboEmbólica (RIETE) and the final model was compared with RIETE and VTE-BLEED scores. The RITE is an ongoing international registry of patients with VTE. Learning was conducted prospectively using data from patients who were recruited from the RIETE registry.

Internal prospective validation was performed with a new cohort of 10,337 patients and external validation was conducted using data from 3027 patients from the COMMAND-VTE database.

Continue Reading

There were 5 ML methods used for training; the best performing method was XGBoost. In the prospective validation cohort, the rate of major bleeding was 2.2%. The algorithm correctly identified 8.7% of major bleeding episodes and failed to identify 1.6% (odds ratio [OR], 5.9; 95% CI, 4.4-7.8). The sensitivity and specificity of the algorithm in the prospective validation cohort was 33.2% and 93%, respectively. The positive predictive value was 10%.

The F1 score was 15.4%, which was higher than the 8.6% with the RIETE score and the 6.4% with the VTE-BLEED score.

“The XGBoost algorithm did not perform well in the external validation cohort,” the authors wrote in their report. They noted that this was because the COMMAND-VTE dataset “lacked 14 predictors in the algorithm.”

In the external validation cohort, the sensitivity and specificity were 10.3% and 87.6%, respectively. The positive predictive value was 3.5%. The F1 score was 5.2% compared with 17.3% with the RIETE score and 9.75% with the VTE-BLEED score.

The authors concluded that the performance of the algorithm was “better than that from the RIETE and VTE-BLEED scores only in the prospective validation cohort, but not in the external validation cohort.”

Disclosures: Some of the study authors declared affiliations with biotech, pharmaceutical, or device companies. Please see the original reference for a full list of disclosures.


Mora D, Mateo J, Nieto JA, et al. Machine learning to predict major bleeding during anticoagulation for venous thromboembolism: possibilities and limitations. Br J Haematol. Published online March 21, 2023. doi: 10.1111/bjh.18737