Abstract WMP80: Multimodal CT Imaging Of Ischemic Stroke Thrombi Identifies Scale-Invariant Radiomic Features

• DOI: 10.1161/str.54.suppl_1.wmp80
• Publication Date: 2023-02-02T10:16:12Z
• Authors: Briana Santo; S M M J Sarayi; Andre Monteiro; Adnan H Siddiqui; Vincent Tutino
• Keywords: Stroke

Background: Paired analysis of ischemic stroke clot histology and pre-treatment CT imaging could identify digital indicators treatment outcome or etiology. Yet, differences in scale limit our ability to interpret the biology from images. Hypothesis: Radiomic resected clots on microCT has potential important, biologically-interpretable features that are correlated among modalities (CTA, nCCT) post-treatment histopathology. Methods: We performed multimodal 10 retrieved by mechanical thrombectomy (example top Figure). Clots were manually segmented co-registered, CTA nCCT For same cases, portions stained with iodine, imaged a ScanCo 100 (4.9 μm resolution), using adaptive intensity thresholding. texture ( n =93 for each modality, 279 total) extracted images PyRadiomics. Pearson correlation was used test associations between (or radiomics. Statistical significance considered at R ≥0.7 q <0.05, after multiple testing correction. Results: 18 significantly microCT, 5 (bottom Cluster Prominence =0.96, <0.001) Shade =0.92, exhibited near-perfect positive correlation. When comparing nCCT, FOS Energy =0.90, Total had high Our findings suggest RBC density (reflected brighter pixels) heterogeneity constituents pixel intensity) scale-invariant attributes detectable CT. Conclusion: Scale-invariant radiomic can help bridge gap pathobiology.