Author(s) :

Dr.P.Saravanan.

Volume/Issue :

Abstract :

Breast cancer remains one of the leading causes of mortality among women worldwide, underscoring the urgent need for early and accurate diagnosis. While Convolutional Neural Networks (CNNs) have been extensively used in medical image analysis for breast cancer detection, their performance is highly sensitive to network architecture and hyperparameter selection. This paper introduces a novel hybrid framework that integrates Brain Storm Optimization (BSO) with CNNs to enhance diagnostic accuracy and optimize treatment planning. By employing BSO—a population-based optimization algorithm inspired by human brainstorming—we automatically tune CNN parameters for feature extraction and classification tasks on mammography and ultrasound datasets. The experimental results indicate that our CNN-BSO hybrid model outperforms traditional CNN approaches, yielding higher accuracy, sensitivity, and specificity. This work not only demonstrates significant improvements in breast cancer identification but also paves the way for more robust computer-aided diagnostic systems.

No. of Downloads :

0