Executive Summary : | Stroke represents a debilitating condition that can lead to functional dependence and death. Hence, timely and accurate detection of the extent of salvageable tissue in acute ischemic stroke is crucial. Several modern neuroimaging modalities have been utilized to address this challenge. These include different magnetic resonance techniques - diffusion-weighted imaging, perfusion imaging and arterial spin labeling. However, these modalities are either limited in their capacity or have potential side effects. Recent literature suggests that resting-state functional MRI (rs-fMRI) is a safe, non-invasive and effective alternative to assess changes in cerebral hemodynamics caused by stroke. They suggest that temporal shifts observed in rs-fMRI represent spatial differences in arterial arrival times due to cerebrovascular diseases. Specifically, areas of abnormal perfusion depict time shift in low-frequency signals compared to global mean signal similar to the delayed time-to-peak in dynamic susceptibility-weighted contrast-enhanced perfusion imaging. Although quite promising, the time-shift analysis technique requires comprehensive validation before clinical translation. While several studies using time-shift analysis demonstrated high spatial correspondence with areas of hypo-perfusion as detected by perfusion imaging in individuals with acute stroke, they have been limited to a small number of samples. Hence, further validation is required in larger cohorts. Furthermore, sensitivity of TSA to factors such as motion artifacts needs to be explored further. In addition to time-to-peak, correlation to other perfusion parameters such as mean transient time might provide more insights. Finally, the influence of scanning time on the quality of results needs to be investigated thoroughly. In addition to correlation based time-shift analysis, several other BOLD signal metrics have been identified in literature that are related to perfusion-related changes. These include coefficient of variation, amplitude of low frequency fluctuations (ALFF) and regional homogeneity (ReHo). Furthermore, while the metrics suggested in literature are effective, they only capture simple, linear trends in the signals. However, comprehensive comparison of these metrics for detecting salvageable brain parenchyma has not been conducted. Inspired by these studies, this project aims to use an interdisciplinary approach to developing a neuroimaging tool based on rs-fMRI BOLD signal to characterize the salvageable brain parenchyma in stroke patients. Systematic evaluation of reliability and stability of the proposed technique will be conducted with the aim of introducing a novel diagnostic tool for improved selection of stroke patients. Insights gained from this novel diagnostic tool will allow for timely implementation of already effective interventions in a cost-effective manner. |
Co-PI: | Dr. Balwinder Singh Sodhi, Indian Institute Of Technology (IIT) Ropar, Punjab,Nangal Road, Hussainpur,Punjab,Rupnagar-140001, Dr. Jitender Saini, National Institute Of Mental Health And Neurosciences, Bangalore, Karnataka-560029 |