Authors:
B. Karthikeyan, R. Ashika Devi, Mita Munshi
Addresses:
Department of Computer Science, Bishop Heber College, Trichy, Tamil Nadu, India, bkarthikeyanphd@gmail.com, ashikadevi2002@gmail.com. Department of Food Science and Human Nutrition, Iowa State University, Ames, Iowa, United States of America, mmunshi@iastate.edu.
Brain MRI segmentation affects study outcomes in numerous clinical applications. Several processing pathways require correct anatomical area division. Magnetic resonance imaging (MRI) for brain anatomy and injury detection has improved. Advanced brain MR imaging has provided more high-quality data. Clinicians struggle to manually extract crucial information from these huge and complex MRI datasets. Many inter- or intra-operator variability studies make manual examination error-prone and time-consuming. Computerized techniques are needed to improve brain MRI data interpretation and disease detection and diagnosis. The MRI scans exhibit few artifacts and strong tissue contrast. The great contrast between soft tissues is one of its many advantages over other imaging methods. MRI data is too large for human processing, which has hindered its application. Image pre-processing, feature extraction, segmentation, and classification are needed for MRI tumour detection. This research can use convolutional neural networks and other image segmentation approaches. Deep neural network algorithms can determine a person’s health at the final stage. Medical imaging has achieved reproducible segmentation, anomaly categorization, and disease picture intensities with improved accuracy and reduced error rates.
Keywords: Magnetic Resonance Imaging (MRI); Input MR Image; Confusion Matrix; MRI Datasets; Generally Speaking; Convolutional Neural Networks; Artificial Intelligence (AI); Machine Learning (ML).
Received on: 05/08/2023, Revised on: 12/10/2023, Accepted on: 01/11/2023, Published on: 07/03/2024
AVE Trends in Intelligent Health Letters, 2024 Vol. 1 No. 1, Pages: 1-9