Ischemic Brain Stroke Segmentation Using Various Attention Mechanism Coupled with Residual Block

Ischemic stroke, caused by cerebral blood flow obstruction, remains a critical health concern with potentially permanent disabilities. Automated lesion segmentation plays a vital role in clinical diagnostics, where rapid and precise identification of affected brain regions is crucial for optimal treatment outcomes. The exhibition of these models was assessed on 94 patient cases utilizing the ISLES 2018 dataset, which incorporates an assortment of CT pictures, including standard CT, CT_4DPWI, and different blood stream related maps (CBF, CBV, MTT, and Tmax).