Abstract:

Breast cancer is one of the most prevalent and life-threatening diseases affecting women worldwide. Early and accurate diagnosis is crucial for improving survival rates and treatment outcomes. This study aims to investigate the feasibility of using logistic regression analysis to differentiate between malignant and benign breast tumors based on various clinical and imaging features.The dataset used in this research consists of a comprehensive collection of patient data, including age, tumor size, tumor shape, margin, and other relevant attributes, along with corresponding diagnostic outcomes (malignant or benign). Logistic regression, a widely used statistical technique for binary classification, is applied to model the relationship between these features and the tumor classification. The CNN algorithm, known for its prowess in image recognition tasks, is then applied to extract intricate patterns and features from mammographic images. This deep learning model enhances the sensitivity and specificity of the detection process, capturing subtle nuances that may escape traditional methods. The proposed hybrid model synergistically combines the strengths of LR and CNN, resulting in a comprehensive and accurate breast cancer detection system. The integration of LR aids in efficient preprocessing, reducing computational complexity for the CNN. Experimental results on benchmark datasets demonstrate the superior performance of the hybrid model, showcasing its potential as an effective tool for early and accurate breast cancer diagnosis

Description:

Breast Cancer Detection, C, Classification, deep learning, logistic regression