A novel breast cancer diagnostic using convolutional squared deviation neural network classifier with Al-Biruni Earth Radius optimization in medical IoT system

Abstract

Accurate and effective breast cancer diagnosis is crucial for breast cancer early rehabilitation and treatment in the IoT medical environment. Life has changed dramatically for the Internet of Things over the past few years as a result of the development of artificial intelligence and data mining technologies, which offer methods for analyzing both current and past data. In this study, we present an IoT-based medical diagnosis system that can successfully discriminate malignant individuals from positive individuals in an IoT environment to address the challenge of early breast cancer detection. An innovative optimization technique built on the Advanced Al-Biruni Earth Radius (ABER) optimization algorithm improved the classification of breast cancer cases. We suggest semantic picture segmentation of breast cancer histology in this article. The enhanced U-Net architecture for map partitioning is partitioned concurrently. Then, regions of interest are extracted using segmentation, and morphological and texture features are computed. A Convolutional Squared Deviation Neural Network Classifier (CSDNN) classifies tumors into six groups based on specific criteria. Using the Wisconsin Breast Cancer Diagnosis (WDBC) dataset, we evaluated the suggested methodology. A series of simulations was run to show the ABER-CSDNN method's superior performance, and the results reveal promising performance when compared to the most recent state-of-the-art techniques. Accuracy of proposed method achieves 99.12%.

Keywords

• Medical IoT
• Convolutional Squared Deviation Neural Network
• Al-Biruni Earth Radius (ABER) optimization