This study aims to segment the final stroke infarct using acute stroke data. The proposed framework uses nnU-Net, one of the popular deep neural network-based algorithms for segmentation tasks. The neural network algorithm was trained using the datasets shared as a part of the ISLES’24 challenge hosted by MICCAI 2024. The performance of the trained neural network is evaluated using the Dice score, the absolute volume difference, the absolute lesion count difference, and lesion-wise F1 score. In total, seven different options for neural network training were performed and the best results were achieved through data augmentation techniques and by using only computed tomography images.