Date created: | Last Updated:

Category: Analysis

Description: Stroke is a leading cause of disability and mortality, resulting in substantial socio-economic burden for healthcare systems. With advances in artificial intelligence, visual image information can be processed into numerous quantitative features in an objective, repeatable and high-throughput fashion, in a process known as radiomics analysis (RA). Recently, investigators have attempted to apply RA to stroke neuroimaging in the hope of promoting personalized precision medicine. This review aimed to evaluate the role of RA in acute ischemic stroke (AIS) outcome prediction. We conducted a systematic review following the PRISMA guidelines, searching PubMed and Embase using the keywords: ‘magnetic resonance imaging (MRI)’, ‘radiomics’, and ‘stroke’. The PROBAST tool was used to assess the risk of bias. Also, radiomics quality score (RQS) was applied to evaluate the methodological quality of radiomics studies. From the 150 abstracts returned by electronic lit-erature research, 6 studies fulfilled the inclusion criteria. Five studies evaluated predictive value for different predictive models (PMs). In all studies, the combined PMs consisting of clinical and ra-diomics features have achieved the best predictive performance compared to PMs based only on clinical or only on radiomics features, the results varying from an AUC of 0.80 (95% CI, 0.75-0.86) to an AUC of 0.92 (95% CI, 0.87-0.97). Two studies developed a radiomics- and clinical-based nom-ogram, the best predictive performance was achieved by the one based on more clinical factors and less, but better selected radiomics features. The median RQS of the included studies is 15, reflecting a moderate methodological quality. Assessing the risk of bias using PROBAST, potential high risk of bias in participants selection was identified. Our findings suggest that combined PMs integrating both clinical and radiomics variables seems to achieve best predictive results. Although, radiomics studies findings are significant in research field, these PMs should be validated in multiple clinical settings to reduce potential selection risk of bias and to become relevant prognosis tools in clinical practice.