Design and development of an effective classifier for medical images based on machine learning and image segmentation

Recently, there has been an increase in the death rate due to encephaloma tumours affecting all age groups. Because of their intricate designs and the interference they cause in diagnostic imaging, these tumours are notoriously difficult to spot. Early and accurate detection of tumours is crucial because it allows for identifying and predicting malignant regions using medical imaging. Using segmentation and relegation techniques, medical scans can aid clinicians in making an early diagnosis and potentially save time. On the other hand, the identification of tumours may be a laborious and extended process for professional doctors owing to the complex nature of tumour formations and the presence of noise in the data produced by Magnetic Resonance Imaging (MRI) since it is pretty imperative to locate and determine the site of the tumour as quickly as feasible. This research proposes a method for detecting brain cancers from MRI scans based on machine learning. It uses the Support Vector Machine, K Nearest Neighbor, and Nave Bayes algorithms for image preprocessing, picture segmentation, feature extraction, and classification. According to the findings, the SVM algorithm accomplished the best level of accuracy, which is 89 %.