RESUMO
OBJECTIVES: The Alberta Stroke Program Early CT Score (ASPECTS), a systematic method for assessing ischemic changes in acute ischemic stroke using non-contrast computed tomography (NCCT), is often interpreted relying on expert experience and can vary between readers. This study aimed to develop a clinically applicable automatic ASPECTS system employing deep learning (DL). METHODS: This study enrolled 1987 NCCT scans that were retrospectively collected from four centers between January 2017 and October 2021. A DL-based system for automated ASPECTS assessment was trained on a development cohort (N = 1767) and validated on an independent test cohort (N = 220). The consensus of experienced physicians was regarded as a reference standard. The validity and reliability of the proposed system were assessed against physicians' readings. A real-world prospective application study with 13,399 patients was used for system validation in clinical contexts. RESULTS: The DL-based system achieved an area under the receiver operating characteristic curve (AUC) of 84.97% and an intraclass correlation coefficient (ICC) of 0.84 for overall-level analysis on the test cohort. The system's diagnostic sensitivity was 94.61% for patients with dichotomized ASPECTS at a threshold of ≥ 6, with substantial agreement (ICC = 0.65) with expert ratings. Combining the system with physicians improved AUC from 67.43 to 89.76%, reducing diagnosis time from 130.6 ± 66.3 s to 33.3 ± 8.3 s (p < 0.001). During the application in clinical contexts, 94.0% (12,591) of scans successfully processed by the system were utilized by clinicians, and 96% of physicians acknowledged significant improvement in work efficiency. CONCLUSION: The proposed DL-based system could accurately and rapidly determine ASPECTS, which might facilitate clinical workflow for early intervention. CLINICAL RELEVANCE STATEMENT: The deep learning-based automated ASPECTS evaluation system can accurately and rapidly determine ASPECTS for early intervention in clinical workflows, reducing processing time for physicians by 74.8%, but still requires validation by physicians when in clinical applications. KEY POINTS: The deep learning-based system for ASPECTS quantification has been shown to be non-inferior to expert-rated ASPECTS. This system improved the consistency of ASPECTS evaluation and reduced processing time to 33.3 seconds per scan. 94.0% of scans successfully processed by the system were utilized by clinicians during the prospective clinical application.
RESUMO
Currently, clinically available coronary CT angiography (CCTA) derived fractional flow reserve (CT-FFR) is time-consuming and complex. We propose a novel artificial intelligence-based fully-automated, on-site CT-FFR technology, which combines the automated coronary plaque segmentation and luminal extraction model with reduced order 3 dimentional (3D) computational fluid dynamics. A total of 463 consecutive patients with 600 vessels from the updated China CT-FFR study in Cohort 1 undergoing both CCTA and invasive fractional flow reserve (FFR) within 90 d were collected for diagnostic performance evaluation. For Cohort 2, a total of 901 chronic coronary syndromes patients with index CT-FFR and clinical outcomes at 3-year follow-up were retrospectively analyzed. In Cohort 3, the association between index CT-FFR from triple-rule-out CTA and major adverse cardiac events in patients with acute chest pain from the emergency department was further evaluated. The diagnostic accuracy of this CT-FFR in Cohort 1 was 0.82 with an area under the curve of 0.82 on a per-patient level. Compared with the manually dependent CT-FFR techniques, the operation time of this technique was substantially shortened by 3 times and the number of clicks from about 60 to 1. This CT-FFR technique has a highly successful (> 99%) calculation rate and also provides superior prediction value for major adverse cardiac events than CCTA alone both in patients with chronic coronary syndromes and acute chest pain. Thus, the novel artificial intelligence-based fully automated, on-site CT-FFR technique can function as an objective and convenient tool for coronary stenosis functional evaluation in the real-world clinical setting.