Diagnosing Breast Cancer with A.I. | Meyyammai Meyyappan & Sabari Laxmi | TEDxPOWIIS Youth

Breast Cancer Diagnosis & AI Application - Breast cancer is highly curable if detected early, but traditional diagnostics can be lengthy, costly, and subjective [3][4] - The project utilizes machine learning to analyze cell morphology for breast cancer diagnosis, training AI to differentiate between benign and malignant cells [4] - AI studies correlations between data, using 30 different variables to predict tumor malignancy [15][18] - The model uses logistic regression, support vector machine, random forest, and gradient boosting to improve prediction accuracy [19][20] Technical Details & Data - The model is trained using the Wisconsin breast cancer data set, which includes 569 samples across 30 key features and 10 cell attributes [12] - Key cell morphology features include radius, area, texture, and perimeter, which are altered in cancerous cells due to excess DNA activity [9][10][11] - The machine learning model provides specific measurements, such as radius (7.5%) and area (920), to indicate malignancy [9] AI in Medicine & Future Implications - AI can provide a second opinion by analyzing numerical data, potentially catching what might be missed by human pathologists [25][26] - The project aims to assist doctors and pathologists, not replace them, by providing a second pair of eyes trained on vast amounts of data [26] - AI's capabilities and limitations must be understood to fully utilize its potential in medicine and other industries [25]