Related Experiment Video
Updated: Dec 12, 2025
09:59
A Magnetic Resonance Imaging Protocol for Stroke Onset Time Estimation in Permanent Cerebral Ischemia
Published on: September 16, 2017
14.5K
Apparent diffusion coefficient map based radiomics model in identifying the ischemic penumbra in acute ischemic
Ru Zhang1, Li Zhu1, Zhengqi Zhu1
1Department of Radiology, The Second Affiliated Hospital of Nantong University, Nantong, China.
Annals of Palliative Medicine
|August 14, 2020
Summary
This study shows that an apparent diffusion coefficient (ADC) map radiomics model can effectively identify the ischemic penumbra (IP) in acute ischemic stroke (AIS) patients, aiding in treatment decisions.
Area of Science:
- Radiology
- Medical Imaging Analysis
- Stroke Research
Background:
- Acute ischemic stroke (AIS) treatment hinges on preserving the ischemic penumbra (IP).
- Radiomics analysis of apparent diffusion coefficient (ADC) maps offers a novel approach for IP identification.
Purpose of the Study:
- To evaluate the efficacy of an ADC map-based radiomics model in identifying the IP in AIS patients.
- To assess the model's predictive performance and clinical utility.
Main Methods:
- Retrospective analysis of 241 AIS patients using perfusion-weighted imaging (PWI)/diffusion-weighted imaging (DWI) mismatch as the gold standard.
- Extraction of 896 features from ADC maps, followed by feature selection using mRMR and LASSO algorithms.
- Model validation through ROC curves, internal cross-validation, and decision curve analysis (DCA).
Main Results:
- A radiomics model comprising 21 features was developed.
- The model achieved high performance with an AUC of 0.92 in the training set and 0.90 in the validation set.
- Decision curve analysis indicated significant clinical benefit within a specific threshold range.
Conclusions:
- The ADC map-based radiomics model demonstrates effectiveness in identifying the IP in AIS.
- This model holds promise for improving diagnostic accuracy and guiding treatment strategies for AIS.
