<![CDATA[The increasing prevalence of breast cancer, one of the cancers with the highest incidence rate in the world, demands the development of effective early detection methods to increase patients' chances of recovery and reduce treatment costs. The main challenge in early detection of breast cancer is the identification of early symptoms that are often not felt, so many cases are only detected at an advanced stage. This study aims to develop a breast cancer early detection model using ultrasound images based on the convolutional neural networks (CNN) artificial neural network method optimized with the particle swarm optimization (PSO) algorithm. CNN is used to extract complex features from medical images, while PSO optimizes model parameters to improve accuracy. The dataset consists of 1,607 images classified into normal, benign, and malignant categories, through the process of model training and validation. The results showed that the integration of particle swarm optimization and convolutional neural networks resulted in an accuracy of 90.67%, higher than the convolutional neural networks method alone at 87.33%. In addition, the CNN-PSO model also excels in precision, sensitivity, and F1-score value. This research provides an effective diagnostic technology solution for primary healthcare facilities, with implications for improved early detection and reduced delayed diagnosis. This technology can be applied more widely through training of health workers and equitable distribution of diagnostic devices to improve service accessibility.]]>
Copyrights © 2025
