A pattern classification analysis tool that potentially increased brain tumor diagnostic procedures. By taking an information picture, assign significance to different viewpoints in the picture and classify each case.
Dataset - https://www.kaggle.com/datasets/navoneel/brain-mri-images-for-brain-tumor-detection
In this era of unprecedented changes, with the increase in ailments, the field of health demands the use of new technology to provide solutions to these problems. The brain tumor is one such malady that is the most life-threatening disease known to mankind. Even if this disease is detected in the primal stage, it’s not completely possible to find a cure within the time frame and save the patient’s life.
The Tumor can be classified into two types: Malignant (meaning the cells are cancerous and lethal to the patient’s life) Benign (meaning the presence of the tumor won’t affect the health of the patient in any manner).
It is commonly accepted that the most important part of the human body is the human brain, therefore, if anything shall happen to it, it will directly impact the life expectancy of the patient.
The development of a prediction model for the diagnosis of brain tumors would greatly benefit the medical community.
In this project, a model is built with a combination of a classification problem which is used to predict whether the subject has a brain tumor or not, and a Computer Vision problem which is to automate the process of brain cropping from MRI scans.
The project is divided into 3 parts: data input and preprocessing, building the model, and image classification using the built model.
Data Mining is the discovery of hidden information in databases and can be viewed as a step in the knowledge discovery process. Data mining functions include clustering, classification, prediction, and link analysis (associations).