Explainable Artificial Intelligence, widely known as interpretable artificial intelligence, is the key component of artificial intelligence. It helps to characterize model accuracy, fairness, transparency, and outcomes of a model in decision-making. The core idea behind LIME XAI leads to new insight into how machine learning models work and validates model efficiency. Especially, in the medical field diagnosing a disease with an explanation about the prediction is a milestone in recent years. This work proposes an attention encoderdecoder, skip connection that combines spatial information of MRI images from the down-sampling path with the up-sampling path to retain good spatial information. Soft attention implemented at the skip connections will actively suppress activations at irrelevant regions. The bilinear interpolation is used in the upsampling process to retain even small details of an image that is lost during encoding and decoding.