Penumbra quantification from MR SWI-DWI mismatch and its comparison with MR ASL PWI-DWI mismatch in patients with
Rupsa Bhattacharjee1,2, Rakesh Kumar Gupta3, Biplab Das4,5
1Centre for Biomedical Engineering, Indian Institute of Technology Delhi, New Delhi, India.
Abstract:
In acute-ischemic-stroke patients, penumbra assessment plays a significant role in treatment outcome. MR perfusion-weighted imaging (PWI) and diffusion-weighted imaging (DWI) mismatch ratio can provide penumbra assessment. Recently reported studies have shown the potential of susceptibility-weighted imaging (SWI) in the qualitative assessment of penumbra. We hypothesize that quantitative penumbra assessment using SWI-DWI can provide an alternative to the PWI-DWI approach and this can also reduce the overall scan-time. The purpose of the current study was to develop a framework for accurate quantitative assessment of penumbra using SWI-DWI and its validation with PWI-DWI-based quantification. In the current study, the arterial-spin-labelling (ASL) technique has been used for PWI. This retrospective study included 25 acute-ischemic-stroke patients presenting within 24 hours of the last noted baseline condition of stroke onset. Eleven patients also had follow-up MRI within 48 hours. MRI acquisition comprised DWI, SWI, pseudo-continuous-ASL (pCASL), FLAIR and non-contrast-angiography sequences. A framework was developed for the enhancement of prominent hypo-intense vein signs followed by automatic segmentation of the SWI penumbra ROI. Apparent-diffusion-coefficient (ADC) maps and cerebral-blood-flow (CBF) maps were computed. The infarct core ROI from the ADC map and the ASL penumbra ROI from CBF maps were segmented semiautomatically. The infarct core volume, SWI penumbra volume (SPV) and pCASL penumbra volume were computed and used to calculate mismatch ratios and . The Dice coefficient between the SWI penumbra ROI and ASL penumbra ROI was 0.96 ± 0.07. correlated well (r = 0.90, p < 0.05) with , which validates the hypothesis of accurate penumbra assessment using the SWI-DWI mismatch ratio. Moreover, a significant association between high SPV and the presence of vessel occlusion in the MR angiogram was observed. Follow-up data showed salvation of penumbra tissue (location and volumes predicted by proposed framework) by treatments. Additionally, functional-outcome analysis revealed 93.3% of patients with > 1 benefitted from revascularization therapy. Overall, the proposed automated quantitative assessment of penumbra using the SWI-DWI mismatch ratio performs equivalently to the ASL PWI-DWI mismatch ratio. This approach provides an alternative to the perfusion sequence required for penumbra assessment, which can reduce scan time by 17% for the protocol without a perfusion sequence.
Insights
Quantitative assessment of the ischemic penumbra using susceptibility-weighted imaging (SWI) and diffusion-weighted imaging (DWI) offers an accurate alternative to perfusion-weighted imaging (PWI)-DWI. This novel SWI-DWI approach reduces overall scan time for acute ischemic stroke patients.
Area of Science:
- Neuroimaging
- Radiology
- Stroke Medicine
Background:
- Penumbra assessment is crucial for optimizing acute ischemic stroke treatment outcomes.
- Current methods like perfusion-weighted imaging (PWI) and diffusion-weighted imaging (DWI) mismatch ratio are established but can be time-consuming.
- Susceptibility-weighted imaging (SWI) has shown potential for qualitative penumbra assessment.
Purpose of the Study:
- To develop and validate a quantitative framework for penumbra assessment using SWI-DWI.
- To evaluate SWI-DWI as a potential time-saving alternative to PWI-DWI.
- To correlate SWI-DWI penumbra assessment with clinical outcomes and treatment response.
Main Methods:
- Retrospective analysis of 25 acute ischemic stroke patients within 24 hours of symptom onset.
- MRI acquisition included DWI, SWI, pseudo-continuous arterial spin labeling (pCASL) for PWI, FLAIR, and non-contrast angiography.
- Developed a framework for SWI penumbra region of interest (ROI) segmentation and compared it with semiautomatically segmented ASL penumbra and ADC-derived infarct core ROIs.
Main Results:
- High Dice coefficient (0.96 ± 0.07) between SWI and ASL penumbra ROIs indicates strong agreement.
- SWI-DWI mismatch ratio strongly correlated with ASL PWI-DWI mismatch ratio (r=0.90, p<0.05), validating the hypothesis.
- Higher SWI penumbra volume correlated with vessel occlusion, and follow-up data showed tissue salvage predicted by the SWI-DWI framework; 93.3% of patients with a mismatch ratio >1 benefited from revascularization.
Conclusions:
- Automated quantitative penumbra assessment using SWI-DWI is equivalent to ASL PWI-DWI.
- This SWI-DWI method offers a viable alternative to perfusion imaging, potentially reducing scan time by 17%.
- The SWI-DWI approach facilitates accurate penumbra assessment, aiding treatment decisions and predicting therapeutic benefits in acute ischemic stroke.
