Author(s): Greeshma Madhukumari

Abstract: The brain is the most complex organ in the human body. Brain Stroke is a long-term disability disease that occurs all over the world and is the leading cause of death. A stroke occurs when the brain’s blood supply is cut off and it ceases to function. Nevertheless, none of the traditional mechanisms focused on finding Brain Stroke along with Risk Factor Identification by combining Clinical Data and CT Image Data. Therefore, this paper proposes an innovative model named XAI-SDCNN-based brain stroke detection and risk factor identification using EBtrapP-FUZZY. Initially, the input clinical dataset is gathered and pre-processed for missing value imputation and one hot encoding, thus improving the data quality. Next, the data is augmented via the proposed QuanGutileSMOTE. Thereafter, the Clinical Features are extracted and will be given to the XAI-SDCNN. If the classified result is a stroke, then the corresponding CT brain image is taken and pre-processed for Noise removal and sharpening. Next, the CT image data is augmented via the geometric transformations. Thereafter, the lesion voxels are segmented using REWS-D2KMC. Next, from the segmented image, features are extracted, and these extracted features are given to REWSO for feature selection. After that, the reduced features are given to the XAI-SDCNN classifier to classify the different types of brain stroke. Finally, EBtrapP-FUZZY-based rules are generated to classify the risk factors.

DOI: 10.33545/27076571.2025.v6.i2c.204

Pages: 248-253 | Views: 90 | Downloads: 31

Download Full Article: Click Here