Detection and Classification of Breast Cancer from Microscopic Biopsy Images using Modified Neural Network

• Published in: 2022 International Conference on Automation, Computing and Renewable Systems (ICACRS) pp. 1259 - 1265
• Main Authors: Vanitha, L., Jayamani, K., Sasirekha, N., V, Yogalakshmi, G, Sajiv
• Format: Conference Proceeding
• Language: English
• Published: IEEE 13.12.2022
• Subjects: Biological system modeling; Biopsy; improvement of estimation techniques; Medical services; Microscopy; Neural networks; pre-diagnosis models; Renewable energy sources; Sensitivity
• DOI: 10.1109/ICACRS55517.2022.10029134

Cancer is widely recognized as being among the most lethal illnesses currently affecting humans. Cancer, which has life-threatening effects, has the opportunity of being eradicated if it is caught at a preliminary phase and treated appropriately. In contrast to that, the precision with which one makes predictions is an important factor. As a result, the development of a dependable model that makes significant contributions to the healthcare community in the prompt identification of biopsy pictures with 100% accuracy becomes the primary focus of attention. In the context of clinical cancer research, the purpose of this article is to work toward the development of improved prediction models by utilizing multivariate information and high-resolution diagnostic techniques. In order to categorize microscopic biopsy pictures of cancer, the socially spider optimization (SSO) encryption method neural network is presented as a possible solution in this research. The relevance of the approach that was developed is dependent on the SSO method's ability to effectively tune the weights assigned to the neural network classification. The performance of the proposed technique is evaluated using measurement systems like accuracy, sensitivity, specificity, as well as MCC assessments, and the conclusions drawn are 95.91 percent, 94.25 percent, 97.12 percent, and 97.68 percent, correspondingly. These results demonstrate that the proposed technique is efficient for the diagnosis of cancer disease.