Performance analysis of machine learning algorithm of detection and classification of brain tumor using computer vision

• Published in: Advances in engineering software (1992) Vol. 173; p. 103221
• Main Authors: Shinde, Ashwini S, Mahendra, BM, Nejakar, Santosh, Herur, Santosh M, Bhat, Nagaraj
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-11-2022
• Subjects: Artificial Intelligence; Logistic regression; MRI; Multilayer Perceptron; OASIS; Logistic regression; OASIS; MRI; Artificial Intelligence; Multilayer Perceptron
• ISSN: 0965-9978
• DOI: 10.1016/j.advengsoft.2022.103221

•Brain tumor is one of the undesirables, uncontrolled growth of cells in all age groups.•Classification of tumors depends no its origin and degree of its aggressiveness, it also helps the physician for proper diagnosis and treatment plan.•This research demonstrates the analysis of various state-of-art techniques in Machine Learning such as Logistic, Multilayer Perceptron, Decision Tree.•The research also reveals that the Logistic Regression and the Multilayer Perceptron gives the highest accuracy of 90%.•It mimics the human reasoning that learns, memorizes and is capable of reasoning and performing parallel computations.
Brain tumor is one of the undesirables, uncontrolled growth of cells in all age groups. Classification of tumors depends no its origin and degree of its aggressiveness, it also helps the physician for proper diagnosis and treatment plan. This research demonstrates the analysis of various state-of-art techniques in Machine Learning such as Logistic, Multilayer Perceptron, Decision Tree, Naive Bayes classifier and Support Vector Machine for classification of tumors as Benign and Malignant and the Discreet wavelet transform for feature extraction on the synthetic data that is available data on the internet source OASIS and ADNI. The research also reveals that the Logistic Regression and the Multilayer Perceptron gives the highest accuracy of 90%. It mimics the human reasoning that learns, memorizes and is capable of reasoning and performing parallel computations. In future many more AI techniques can be trained to classify the multimodal MRI Brain scan to more than two classes of tumors.