BreastCancerNet: Flask-Enabled Attention-Driven Hybrid Dual DNN Framework for Real-Time Breast Cancer Prediction.

Area of Science:

• Oncology
• Medical Informatics
• Artificial Intelligence

Background:

• Breast cancer is the most common cancer in women globally, necessitating advanced diagnostic tools.
• Artificial intelligence (AI) shows potential in improving diagnostic accuracy and treatment optimization for cancer.
• Early and accurate detection significantly impacts patient survival rates in breast cancer cases.

Purpose of the Study:

• To develop and evaluate a hybrid AI architecture, BreastCancerNet, for enhanced breast cancer diagnosis.
• To improve the accuracy of differentiating between malignant and benign breast cancer cases.
• To create an interactive web application for real-time breast cancer detection using AI.

Main Methods:

• Utilized the Wisconsin Breast Cancer dataset (569 patients, 30 attributes).
• Developed BreastCancerNet, a hybrid AI model combining dual deep neural networks (DNNs) and an attention mechanism.
• Integrated a Support Vector Machine (SVM) for final classification.

Main Results:

• Achieved a diagnostic accuracy of 99.42% in distinguishing malignant from benign breast cancer.
• The dual DNNs extracted diverse features, while the attention mechanism prioritized critical data points.
• A user-centric web application was developed for real-time, interactive breast cancer detection.

Conclusions:

• BreastCancerNet demonstrates high efficacy in breast cancer diagnosis, achieving near-perfect accuracy.
• The hybrid AI approach effectively leverages deep learning and attention mechanisms for improved diagnostic performance.
• The developed web application offers a practical tool for enhanced breast cancer screening and patient engagement.