Identifying accuracy-preserving truncation thresholds of the time-attenuation curve in CT perfusion for acute anterior circulation large-vessel occlusion stroke.
Authors
Affiliations (3)
Affiliations (3)
- Department of Radiology, the Second Affiliated Hospital of Chongqing Medical University, Chongqing, China.
- Department of Radiology, Jiuzhaigou County Hospital of Traditional Chinese and Tibetan Medicine, Sichuan, China.
- Department of Radiology, the Second Affiliated Hospital of Chongqing Medical University, Chongqing, China. [email protected].
Abstract
To quantify the impact of progressively increasing time-attenuation curve (TAC) truncation on CT perfusion (CTP) parameters in acute anterior circulation large-vessel occlusion (AC-LVO) stroke, and to identify truncation thresholds that preserve quantitative accuracy and the reliability of treatment selection. This retrospective study included 113 consecutive patients with acute AC-LVO stroke. CTP datasets acquired at 1.5- or 2.0-second sampling intervals were processed using two fully automated AI-based software packages (CTPDoc and United Imaging). TAC truncation was simulated by sequential removal of terminal frames from full-scan datasets. Using full-scan maps as the reference standard, quantitative deviations were analyzed with Friedman tests and intraclass correlation coefficients (ICCs). Qualitative agreement regarding treatment eligibility (based on DEFUSE 3, TRACE-III, and SELECT Late criteria) was assessed using chi-square or Fisher's exact tests and percent agreement. Truncation thresholds were defined as the latest time points showing no significant differences (p > 0.05) and good agreement (ICC ≥ 0.75; percent agreement ≥ 82%). Increasing TAC truncation increased parameter deviations and reduced agreement in treatment selection across four groups. Relative to standardized onset, truncation thresholds were 27 and 28 s for CTPDoc (1.5-second and 2.0-second intervals), and 34.5 and 40 s for United Imaging (1.5-second and 2.0-second intervals). Although TAC truncation degrades CTP accuracy and treatment-selection concordance, algorithm- and protocol-specific truncation thresholds exist. Adherence to these thresholds may support reliable treatment selection and reduce unnecessary repeat CTP examinations, thereby preventing treatment delays, cumulative radiation exposure, and contrast load when TACs are partially truncated.