This project aims to develop a machine learning-based system to detect breast cancer types (Benign, Malignant, or Normal) from medical images. The system leverages various classification algorithms, including Logistic Regression, Support Vector Machines (SVM), Random Forest, Decision Trees, Naive Bayes, and K-Nearest Neighbors (KNN). This project was developed as part of my summer internship at IGCAR, where I explored various image classification techniques in medical imaging.
Thermal images captured by the dynamic protocol were used after the cooling ofthe breasts by air stream, 20 sequential images with intervals of 15 seconds between them were taken during the process of returning the patient’s body to thermal equilibrium with the enviroment. The images are stored in the Database for Research Mastology with Infrared Image - DMR-IR, accessible on the website http://visual.ic.uff.br/dmi.
Camera used to capture: FLIR SC-620 Resolution: 640 x 480 pixels
For our project 92 Benign, 42 Malignant, and 38 Normal images were used and 10 different images were tested using 6 different machine learning models. You can find the dataset here
📍 SVM achieved the highest accuracy (77.14%) and a balanced precision-recall performance.
📍 Logistic Regression and Random Forest showed consistent classifications for most Malignant cases. Decision Tree performed poorly with higher variance in classifications.
Languages used :
📣 Feel free to have a look at all the files in this repository !🤗
❎ In case you find issues in any of my Repositories, you can Hit Me Up here! 👈