Related Experiment Video
Updated: Aug 12, 2025
A Magnetic Resonance Imaging Protocol for Stroke Onset Time Estimation in Permanent Cerebral Ischemia
Published on: September 16, 2017
Computed Tomography Angiography-Based Thrombus Radiomics for Predicting the Time Since Stroke Onset
Yue Cheng1, Sunli Wan2, Wenjuan Wu2
1Department of Radiology, Wuxi No.2 People's Hospital, Jiangnan University Medical Center, 68 Zhongshan Road, Wuxi, Jiangsu, China; Department of Radiology, Wuxi NO.2 People's Hospital, Affiliated Wuxi Clinical College of Nantong University, Wuxi, China.
Computed tomography angiography (CTA) radiomics features can estimate time since stroke onset in acute ischemic stroke (AIS) patients. Adding clinical data did not significantly improve the predictive performance of these radiomics models.
Area of Science:
- Radiology
- Medical Imaging
- Artificial Intelligence in Medicine
Background:
- Accurate estimation of time since stroke onset (TSS) is critical for acute ischemic stroke (AIS) treatment decisions.
- Computed tomography angiography (CTA) is a key imaging modality in AIS management.
Purpose of the Study:
- To evaluate the utility of radiomics features (RFs) extracted from CTA images to estimate TSS in AIS patients.
- To compare the performance of a radiomics-only model with a combined model incorporating clinical factors.
Main Methods:
- Retrospective analysis of 221 AIS patients.
- Extraction of 944 radiomics features from CTA-derived thrombus segmentations.
- Development and validation of radiomics and combined clinical-radiomics models.
- Assessment of model performance using Area Under the Receiver Operating Characteristic Curve (AUC) and decision curve analysis.
Main Results:
- The radiomics model achieved an AUC of 0.803 in both training and test cohorts.
- The combined model (including age, diabetes, atrial fibrillation) showed similar AUC values (0.813 training, 0.803 test).
- No significant difference in performance was observed between the radiomics and combined models; both demonstrated clinical utility.
Conclusions:
- CTA-based thrombus radiomics is a viable method for estimating TSS in AIS.
- Incorporating clinical data does not substantially enhance the predictive accuracy of radiomics models for TSS estimation in this cohort.
