Dataset Information

Machine Learning Prediction of Stroke Mechanism in Embolic Strokes of Undetermined Source.

ABSTRACT:

Background and purpose

One-fifth of ischemic strokes are embolic strokes of undetermined source (ESUS). Their theoretical causes can be classified as cardioembolic versus noncardioembolic. This distinction has important implications, but the categories' proportions are unknown.

Methods

Using data from the Cornell Acute Stroke Academic Registry, we trained a machine-learning algorithm to distinguish cardioembolic versus non-cardioembolic strokes, then applied the algorithm to ESUS cases to determine the predicted proportion with an occult cardioembolic source. A panel of neurologists adjudicated stroke etiologies using standard criteria. We trained a machine learning classifier using data on demographics, comorbidities, vitals, laboratory results, and echocardiograms. An ensemble predictive method including L1 regularization, gradient-boosted decision tree ensemble (XGBoost), random forests, and multivariate adaptive splines was used. Random search and cross-validation were used to tune hyperparameters. Model performance was assessed using cross-validation among cases of known etiology. We applied the final algorithm to an independent set of ESUS cases to determine the predicted mechanism (cardioembolic or not). To assess our classifier's validity, we correlated the predicted probability of a cardioembolic source with the eventual post-ESUS diagnosis of atrial fibrillation.

Results

Among 1083 strokes with known etiologies, our classifier distinguished cardioembolic versus noncardioembolic cases with excellent accuracy (area under the curve, 0.85). Applied to 580 ESUS cases, the classifier predicted that 44% (95% credibility interval, 39%-49%) resulted from cardiac embolism. Individual ESUS patients' predicted likelihood of cardiac embolism was associated with eventual atrial fibrillation detection (OR per 10% increase, 1.27 [95% CI, 1.03-1.57]; c-statistic, 0.68 [95% CI, 0.58-0.78]). ESUS patients with high predicted probability of cardiac embolism were older and had more coronary and peripheral vascular disease, lower ejection fractions, larger left atria, lower blood pressures, and higher creatinine levels.

Conclusions

A machine learning estimator that distinguished known cardioembolic versus noncardioembolic strokes indirectly estimated that 44% of ESUS cases were cardioembolic.

SUBMITTER: Kamel H

PROVIDER: S-EPMC8034802 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundFrom a therapeutic viewpoint, it is important to differentiate the underlying causes of embolism in patients with cryptogenic stroke, such as aortic arch atheroma, patent foramen ovale, and paroxysmal atrial fibrillation. We investigated the clinical and radiological characteristics of these 3 common causes of cryptogenic embolism to develop models for decision making in etiologic workups.Methods and resultsA total of 321 consecutive patients with acute infarcts from cryptogenic embolism were included. Patients were divided into 3 groups-aortic arch atheroma (n=40), patent foramen ovale (n=153), and paroxysmal atrial fibrillation (n=128)-based on extensive cardiologic workups. We used a multinomial logistic regression analysis to detect the clinical and diffusion-weighted imaging factors associated with the probability of aortic arch atheroma, patent foramen ovale, and paroxysmal atrial fibrillation. Clinical and radiological features differed among the groups. The patent foramen ovale group had a healthy vascular risk factor profile and showed posterior circulation involvement compared with other groups (P<0.01). In contrast, paroxysmal atrial fibrillation-related strokes had higher initial National Institutes of Health Stroke Scale (NIHSS) scores and larger lesions than the other groups (P<0.001). The aortic arch atheroma group had clinical features similar to those of the paroxysmal atrial fibrillation group but showed small lesions scattered in multiple vascular territories (P<0.001). Multivariate regression analysis revealed that age, initial NIHSS score, lesion size (≥20 mm), multiple (≥3) lesions, and involvement of posterior circulation or multiple vascular territories differentiated the 3 groups (pseudo, R(2)=0.656). The prediction ability of this model was validated in the external validation cohort (n=117, area under the curve 0.78).ConclusionsOur data indicate that patients with cryptogenic embolic stroke show distinct clinical and radiological features depending on the underlying causes.