RESUMEN
AIMS: Atrial fibrillation is associated with important mortality but the usual clinical risk factor based scores only modestly predict mortality. This study aimed to develop machine learning models for the prediction of death occurrence within the year following atrial fibrillation diagnosis and compare predictive ability against usual clinical risk scores. METHODS AND RESULTS: We used a nationwide cohort of 2,435,541 newly diagnosed atrial fibrillation patients seen in French hospitals from 2011 to 2019. Three machine learning models were trained to predict mortality within the first year using a training set (70% of the cohort). The best model was selected to be evaluated and compared with previously published scores on the validation set (30% of the cohort). Discrimination of the best model was evaluated using the C index. Within the first year following atrial fibrillation diagnosis, 342,005 patients (14.4%) died after a period of 83 (SD 98) days (median 37 [10-129]). The best machine learning model selected was a deep neural network with a C index of 0.785 (95% CI, 0.781-0.789) on the validation set. Compared to clinical risk scores, the selected model was superior to the CHA2DS2-VASc and HAS-BLED risk scores and superior to dedicated scores such as Charlson Comorbidity Index and Hospital Frailty Risk Score to predict death within the year following atrial fibrillation diagnosis (C indexes: 0.597; 0.562; 0.643; 0.626 respectively. P < .0001). CONCLUSION: Machine learning algorithms predict early death after atrial fibrillation diagnosis and may help clinicians to better risk stratify atrial fibrillation patients at high risk of mortality.
RESUMEN
BACKGROUND: Targeting ischemic strokes patients at risk of incident atrial fibrillation (AF) for prolonged cardiac monitoring and oral anticoagulation remains a challenge. Clinical risk scores have been developed to predict post-stroke AF with suboptimal performances. Machine learning (ML) models are developing in the field of AF prediction and may be used to discriminate post-stroke patients at risk of new onset AF. This study aimed to evaluate ML models for the prediction of AF and to compare predictive ability to usual clinical scores. METHODS: Based on a French nationwide cohort of 240,459 ischemic stroke patients without AF at baseline from 2009 to 2012, ML models were trained on a train set and the best model was selected to be evaluate on the test set. Discrimination of the best model was evaluated using the C index. We finally compared our best model with previously described clinical scores. RESULTS: During a mean follow-up of 7.9 ± 11.5 months, 14,095 patients (mean age 77.6 ± 10.6; 50.3% female) developed incident AF. After training, the best ML model selected was a deep neural network with a C index of 0.77 (95% CI 0.76-0.78) on the test set. Compared to traditional clinical scores, the selected model was statistically significantly superior to the CHA2DS2-VASc score, Framingham risk score, HAVOC score and C2HEST score (P < 0.0001). The ability to predict AF was improved as shown by net reclassification index increase (P < 0.0001) and decision curve analysis. CONCLUSIONS: ML algorithms predict incident AF post-stroke with a better ability than previously developed clinical scores. AF: atrial fibrillation; DNN: deep neural network; IS: ischemic stroke; KNN: K-nearest neighbors; LR: logistic regression; RFC: random forest classifier; XGBoost: extreme gradient boosting.