The Impact of a Neuroradiologist on the Report of a Real-World CT Perfusion Imaging Map Derived from Artificial Intelligence/Machine Learning–Driven Software
Year: 2025
Type: article
Study design: Observational Study
Abstract: BACKGROUND AND PURPOSE: According to the guideline, CT perfusion should be read and analyzed by using computer-aided software. This study evaluates the efficacy of artificial intelligence (AI)/machine learning-driven software in CTP imaging and the effect of neuroradiologists' interpretatios on these automated results. MATERIALS AND METHODS: We conducted a retrospective, single-center cohort study from June to December 2023 at a comprehensive stroke center. A total of 132 patients suspected of having acute ischemic stroke underwent CTP using AI software. RapidAI was used for the initial analysis, with subsequent validation and adjustments made by experienced neuroradiologists. The rate of CTP marked as "nonreportable," "reportable," and "reportable with correction" by neuroradiologists was recorded. The degree of confidence in the report of basal and angio-CT scans was assessed before and after CTP visualization. Statistical analysis included logistic regression and F1 score assessments to evaluate the predictive accuracy of AI-generated CTP maps. RESULTS: The study found that CTP maps derived from AI software were reportable in 65.2% of cases without artifacts and improved to 87.9% reportable cases when reviewed by neuroradiologists. Key predictive factors for artifact-free CTP maps included motion parameters and the timing of contrast peak distances. There was a significant shift to higher confidence scores of the angiographic phase of CT after the results of CTP. CONCLUSIONS: Neuroradiologists play an indispensable role in improving the reliability of CTP by interpreting and correcting AI-processed maps.
DOI: 10.3174/ajnr.a8979
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W4413424919