The FIND-AF machine learning algorithm demonstrated stronger discrimination for predicting 6-month incident atrial fibrillation (AUROC 0.824) compared to the CHA2DS2-VASc score (AUROC 0.784).
Cohort (n=2,081,139)
Yes
Does the FIND-AF machine learning algorithm improve prediction of incident atrial fibrillation within 6 months compared to CHA2DS2-VASc and C2HEST scores in individuals aged ≥30 years without known AF?
A machine learning algorithm applied to routinely collected primary care data can accurately identify individuals at high risk of short-term incident atrial fibrillation, outperforming traditional clinical risk scores.
Absolute Event Rate: 0.824% vs 0.784%
Objective Atrial fibrillation (AF) screening by age achieves a low yield and misses younger individuals. We aimed to develop an algorithm in nationwide routinely collected primary care data to predict the risk of incident AF within 6 months (Future Innovations in Novel Detection of Atrial Fibrillation (FIND-AF)). Methods We used primary care electronic health record data from individuals aged ≥30 years without known AF in the UK Clinical Practice Research Datalink-GOLD dataset between 2 January 1998 and 30 November 2018, randomly divided into training (80%) and testing (20%) datasets. We trained a random forest classifier using age, sex, ethnicity and comorbidities. Prediction performance was evaluated in the testing dataset with internal bootstrap validation with 200 samples, and compared against the CHA 2 DS 2 -VASc (Congestive heart failure, Hypertension, Age >75 (2 points), Stroke/transient ischaemic attack/thromboembolism (2 points), Vascular disease, Age 65–74, Sex category) and C 2 HEST (Coronary artery disease/Chronic obstructive pulmonary disease (1 point each), Hypertension, Elderly (age ≥75, 2 points), Systolic heart failure, Thyroid disease (hyperthyroidism)) scores. Cox proportional hazard models with competing risk of death were fit for incident longer-term AF between higher and lower FIND-AF-predicted risk. Results Of 2 081 139 individuals in the cohort, 7386 developed AF within 6 months. FIND-AF could be applied to all records. In the testing dataset (n=416 228), discrimination performance was strongest for FIND-AF (area under the receiver operating characteristic curve 0.824, 95% CI 0.814 to 0.834) compared with CHA 2 DS 2 -VASc (0.784, 0.773 to 0.794) and C 2 HEST (0.757, 0.744 to 0.770), and robust by sex and ethnic group. The higher predicted risk cohort, compared with lower predicted risk, had a 20-fold higher 6-month incidence rate for AF and higher long-term hazard for AF (HR 8.75, 95% CI 8.44 to 9.06). Conclusions FIND-AF, a machine learning algorithm applicable at scale in routinely collected primary care data, identifies people at higher risk of short-term AF.
Nadarajah et al. (Thu,) conducted a cohort in Atrial fibrillation (n=2,081,139). FIND-AF algorithm vs. CHA2DS2-VASc and C2HEST scores was evaluated on Discrimination performance for incident AF within 6 months (AUROC) (95% CI 0.814 to 0.834). The FIND-AF machine learning algorithm demonstrated stronger discrimination for predicting 6-month incident atrial fibrillation (AUROC 0.824) compared to the CHA2DS2-VASc score (AUROC 0.784).