Abstract
Analyzing brain tumours is important for prompt diagnosis and efficient patient care. The morphology of tumours, which includes their size, location, texture, and heteromorphic appearance in medical pictures, makes them difficult to analyse. A unique two-phase deep learning-based framework is suggested in this respect to recognise and classify brain cancers in magnetic resonance images (MRIs). A new DTA approach is suggested in the first phase to successfully identify tumour MRI images from healthy persons. DTA are specifically designed and perform well are used to create the deep boosted feature space, which is then fed into the group of machine learning (ML) classifiers. In the second stage, a brand-new hybrid features fusion-based brain tumour classification technique is put forward, one that makes use of both static and dynamic features as well as an ML classifier to classify various tumour kinds. The proposed algorithm, which can recognise the heteromorphic and variable behaviour of different tumours, is where the dynamic characteristics are taken.In this paper, many segmentation algorithms for MRI and PET are reviewed together with their performance evaluations in order to detect brain tumours. There are numerous segmentation methods available for the diagnosis of medical images. The features of the image, such as the capacity to distinguish between similarities and discontinuities, are often used to classify the segmentation techniques. Neuroradiologists have a difficult issue in trying to quickly identify the abnormal region, which is essential in the medical field. In order to overcome this problem, the efficiency of different segmentation procedures is assessed. The segmentation methods considered here are Ant Colony Optimization (ACO), Wavelet Transform (WT), Gradient Vector Flow (GVF), Gray level Co-occurrence matrix (GLCM), and Artificial Bee Colony (ABC). The various performance metrics are used to evaluate the suggested segmentation algorithms. The GVF strategy works better with MRI images, whereas the ABC and ACO approaches perform similarly with PET scans, according to the data acquired.