Diagnostic test accuracy study of a commercially available deep learning algorithm for ischemic lesion detection on brain MRIs in suspected stroke patients from a non-comprehensive stroke center.

PURPOSE: To estimate the ability of a commercially available artificial intelligence (AI) tool to detect acute brain ischemia on Magnetic Resonance Imaging (MRI), compared to an experienced neuroradiologist. METHODS: We retrospectively included 1030 patients with brain MRI, suspected of stroke from January 6th, 2020 to 1st of April 2022, based on these criteria: Age ≥ 18 years, symptoms within four weeks before the scan. The neuroradiologist reinterpreted the MRI scans and subclassified ischemic lesions for reference. We excluded scans with interpretation difficulties due to artifacts or missing sequences. Four MRI scanner models from the same vendor were used. The first 800 patients were included consecutively, remaining enriched for less frequent lesions. The index test was a CE-approved AI tool (Apollo version 2.1.1 by Cerebriu). RESULTS: The final analysis cohort comprised 995 patients (mean age 69 years, 53 % female). A case-based analysis for detecting acute ischemic lesions showed a sensitivity of 89 % (95 % CI: 85 %-91 %) and specificity of 90 % (95 % CI: 87 %-92 %). We found no significant difference in sensitivity or specificity based on sex, age, or comorbidities. Specificity was reduced in cases with DWI artifacts. Multivariate analysis showed that increasing ischemic lesion size and fragmented lesions were independently associated with higher sensitivity, while non-acute lesion ages lowered sensitivity. CONCLUSIONS: The AI tool exhibits high sensitivity and specificity in detecting acute ischemic lesions on MRI compared to an experienced neuroradiologist. While sensitivity depends on the ischemic lesions' characteristics, specificity depends on the image quality.

Long-term outcomes of endovascular simple coiling versus neurosurgical clipping of unruptured intracranial aneurysms: A systematic review and meta-analysis.

BACKGROUND: Previous studies on ruptured intracranial aneurysms have shown favourable long-term outcomes of patients undergoing endovascular coiling compared to neurosurgical clipping. We aimed to evaluate if these results also apply to patients with unruptured intracranial aneurysms (UIAs). METHODS: Embase, PubMed, and Cochrane Library were systematically searched for all studies reporting long-term (≥3 years) follow-up after coiling or clipping of UIAs. Thirteen studies involving 16,622 coiled patients and 13,606 clipped patients were included. Short-term outcome was defined as death ≤30 days after treatment. Long-term outcomes (>3 years) included all-cause mortality, morbidity (defined as modified Rankin Score 3-5 or Glasgow Outcome Score 2-3), cerebrovascular accident, intracerebral haemorrhage, additional repairs, and lost to follow-up. We calculated relative risk (RR), incidence and mortality rates (IR and MR), together with incidence and mortality rate ratio (IRR and MRR). RESULTS: Patients treated with simple coiling had lower short-term mortality than clipped patients (RR = 0.62 (95%CI 0.42-0.91)), but this difference disappeared after long-term follow-up ((MRR) = 0.89 (95%CI: 0.78-1.02). Coiled patients had higher retreatment rates than clipped patients (IRR = 1.70 (95%CI 1.50-1.93)). CONCLUSIONS: This systematic review and meta-analysis reports benefits and drawbacks of simple coiling versus neurosurgical clipping of UIAs. Future studies with longer follow-up time should account for differences in coiling techniques and confounding factors such as size and location of UIAs.