RAPID Aneurysm: Artificial intelligence for unruptured cerebral aneurysm detection on CT angiography

Objectives: Cerebral aneurysms may result in significant mor-bidity and mortality. Identification of these aneurysms on CT Angiography (CTA) studies is critical to guide patient treat-ment. Artificial intelligence platforms to assist with auto-mated aneurysm detection are of high interest. We determined the performance of a semi-automated artificial intelligence software program (RAPID Aneurysm) for the detection of cerebral aneurysms. Materials and Methods: RAPID Aneurysm was used to detect retrospec-tively the presence of cerebral aneurysms in CTA studies per-formed between January 2019 and December 2020. The gold standard was aneurysm presence and location as determined by the consensus of three expert neuroradiologists. Aneu-rysm detection accuracy, sensitivity, specificity, positive pre-dictive value, negative predictive value, and positive and negative likelihood ratios by RAPID Aneurysm were deter-mined. Results: 51 patients (mean age, 56</n>15; 24 women [47.1%]) with a single CTA were included. A total of 60 aneurysms were identified. RAPID Aneurysm had a sensitiv-ity of 0.950 (95% CI: 0.863-0.983), specificity of 1.000 (95% CI: 0.996-1.000), a positive predictive value (PPV) of 1.000 (95% CI: 0.937-1.000), a negative predictive value (NPV) of 0.997 (95% CI: 0.991-0.999), and an accuracy of 0.997 (95% CI: 0.991-0.999) for cerebral aneurysm detection. Conclusions: RAPID Aneurysm is highly accurate for the detection of cerebral aneurysms on CTA.