Publication Date: 2022/03/03
Abstract: Brain tumours are regarded as one of the most dangerous diseases in both children and adults. Brain tumours make up 85 to 90% of all primary Central Nervous System tumours. Brain tumours are classified as benign, malignant, pituitary, or other. Magnetic Resonance Imaging is the most effective technique for detecting brain tumours. The use of automated classification techniques such as Machine Learning and Artificial Intelligence has consistently demonstrated greater accuracy than manual classification. The user can use the proposed system as a Web Application. The patient, doctor or medical practitioners, paramedic etcetera are the users for the system. The proposed system acts like an assistant to the doctor, by detecting brain cancer in MRI images. The user will upload the brain MR Image of the patient concerned. Then the system will be able to predict whether the patient has cancer or not, if the patient has cancer the category will be specified. Experimental results indicate that the proposed approach outperforms other commonly used methods and gives an overall high validation accuracy
Keywords: Machine Learning, Artificial Intelligence, Convolutional Neural Network, Transfer Learning, Brain Tumor.
DOI: https://doi.org/10.5281/zenodo.6324441
PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT22FEB499.pdf
REFERENCES