An Interpretable Artificial Intelligence Model for Breast Cancer Classification

Pakala Bhanu Prakash
M-Tech, Department of Computer Science and
Engineering,
Vemu Institute of Technology, P. Kothakota,
Chittoor district,
Andhra Pradesh-517112, India.
Email Id: bhanulovely31@gmail.com

Mr. K. Naveen
Assistant Professor, M. Tech, Dept of CSE,
Vemu Institute of Technology, P. kothakota,
Chittoor district,
Andhra Pradesh-517112, India
Email Id: karamalanaveen@gmail.com

Abstract— Breast cancer is one of the most common and fatal diseases in women, which indicates the absence of timely early detection techniques and highlights the necessity of the development of the reliable ones. The paper describes a web-based application that will enhancethe accuracy of breast cancer classification by combining  several machine learning (ML) and deep learning (DL)  models. Developed with the Django platform, the site allows users to create an account and log in to insert healthdata to get predictions. The system is classified into two different datasets which include those based on morphology including shape and size and a set of cytological properties of cell samples. The platform uses some of the following models: LSTM, MLP and CNN to process and analyze data by using different methods like structured input analysis and pattern recognition. To increase transparency, the system incorporates explainable AI tools, which allow users to comprehend what features have contributed to the predictions to make them trustful of the outcomes. The models provide a strong performance, where CNN is the best in analyzing images and MLP and LSTM give correct results in analyzing tabular data. This system helps healthcare practitioners and patients to make sound decisions, which helps in diagnosing breast cancer early with a lot of confidence.
Keywords: Breast cancer detection, early diagnosis, CNN, MLP, LSTM, Django application, user input prediction, explainable AI, clinical support, featureinterpretation.