Precision Medicine: Cascaded CNNs in Brain Tumor Diagnosis and Classification

The human brain serves as the principal regulator of the system, manifesting attributes that are comparable to those of the human species. Aberrations in cellular proliferation and mitosis in the brain can result in the formation of brain malignancies, which can further advance into brain tumours. Computer Vision is crucial in the field of human health as it significantly improves accuracy and decreases the need for human judgement. Traditional imaging modalities, such as computed tomography (CT), X-ray, and magnetic resonance imaging (MRI) examinations, are considered to be secure and dependable. The use of MRI, with its ability to identify even the smallest details, is crucial in the detection of brain cancer. This research aims to examine various techniques for detecting brain cancer using MRI scans. The study entails initial procedural stages, which include the utilisation of the bilateral filter (BF) to eradicate noise in MR images. The following phases entail binary segmentation and Convolutional Neural Network (CNN) segmentation algorithms to accurately identify the tumour location. By applying machine learning techniques to training, testing, and validation datasets, the study forecasts the presence of a brain malignancy in a given individual. To assess the ultimate results, performance metrics including precision, sensitivity, and particularity will be applied. The proposed methodology is expected to provide higher performance in comparison to existing methodologies.