Multiparameter MRI ISODATA characterizes ischemic damage
Purpose: We present a tissue characterization method to identify viable and non-viable tissue in cerebral ischemia using multi-parameter MRI. The method utilizes T1, T2, proton density, and diffusion weighted images (T1WI, T2WI, PDWI, DWI, respectively). Image Analysis Approach: After pre-processing (intra-cranial segmentation, non-uniformity correction, and noise suppression), we segment tissues using a self organizing data analysis method (ISODATA) and characterize tissues using Euclidean distance measures to score the regions between 1 and N (N determines characterization resolution). Signature or score 1 is assigned to normal white matter and score N is assigned to CSF. Each lesion zone is assigned a score based on its levels of differences (in terms of multi-parametric MRI) from white matter and CSF. Experimental Methods: Rats were imaged by a 7T MRI system at one of the three time points (acute, 4–8 hrs; sub-acute, 16–24 hrs; and chronic, 48–168 hrs) after MCA occlusion. Then, they were sacrificed and their brains were sliced and prepared for histological studies. MRIs of 18 slices (8 at acute time in 4 rats, 8 at sub-acute time in 4 rats, and 2 at chronic time in 1 rat) were processed and scored. 2 DWI (b=600, b=800), PDWI, T2WI, and T1WI were used and an MRI score between 1 and 100 (N=100) was found for each tissue region. Segmented tissues were mapped onto the histology images and were scored by an experienced pathologist, from 1 to 10. MRI scores were validated using histology scores. To this end, correlation coefficients between the two scores (MRI and histology) and 95% confidence limits were found. Results: The results showed excellent correlations between MRI and histology scores at different time points. Correlation coefficients were 0.90 acutely, 0.82 sub-acutely, and 0.88 overall. The 95% confidence intervals were (0.53,0.98), (0.27,0.97), and (0.68,0.95), respectively. Conclusion: The proposed method accurately characterizes tissue damage in cerebral ischemia based on multi-parameter MRI. It is useful for a variety of applications such as evaluating response to treatment, where volume changes for different zones of stroke over time, e.g., tissue recovery, can be evaluated.