Background
Early identification of cerebral small vessel disease (CSVD) patients with a higher risk of developing cognitive impairment is essential for timely intervention and improvement of patient prognosis. The advancement of medical imaging and computing capabilities provides new methods for early detection of cognitive disorders. Machine learning (ML) has emerged as a promising technique for cognitive impairment in CSVD. This study aims to conduct a thorough meta-analysis and comparison of published ML prediction models for cognitive impairment in patients with CSVD.
Methods
In September 2024, relevant studies were retrieved from four databases: PubMed, Embase, Web of Science, and the Cochrane Library. The Prediction Model Risk of Bias Assessment Tool (PROBAST) was used to assess the risk of bias of the ML models. The random effects model was used for meta-analysis of C-index, while a bivariate mixed-effects model was used to calculate the pooled sensitivity and specificity with their 95% confidence intervals (CIs). In addition, to limit the influence of heterogeneity, we also performed sensitivity analyses, a meta-regression, and subgroup analysis.
Results
Twenty-one prediction models from thirteen studies, involving 3444 patients met criteria for inclusion. The reported C-index ranged from 0.708 to 0.952. The pooled C-index, sensitivity, and specificity were 0.85 (95% CI 0.82–0.87), 0.82 (95% CI 0.77–0.87), and 0.81 (95% CI 0.73–0.87). As one of the most commonly used ML methods, logistic regression achieved a total merged C-index of 0.81, while non logistic regression models performed better with a total merged C-index of 0.86.
Conclusions
ML models holds significant promise in forecasting the risk of cognitive impairment in patients with CSVD. However, future high-quality research that externally validates the algorithm through prospective studies with larger, more diverse cohorts is needed before it can be introduced into clinical practice.
