IoT based healthcare system using fractional dung beetle optimization enabled deep learning for breast cancer classification

Abstract

Breast cancer classification plays a crucial role in healthcare, especially in the diagnosis and monitoring of patients. Traditional methods for classifying breast cancer based on histopathological images often suffer from limited accuracy, which can hinder early detection and treatment. Hence, this paper devises a novel Internet of Things (IoT) based healthcare system using SqueezeNet_Fractional Dung Beetle Optimization (Squeeze_FDBO) for breast cancer detection. Initially, IoT network is simulated, and routing of the histopathological images to the Base Station (BS) is established utilizing FDBO, which is obtained by combining Dung Beetle Optimizer (DBO), and the Fractional Calculus (FC). At BS, breast cancer classification is done, where input is first processed by a bilateral filter. Then, blood cell segmentation is effectuated using LadderNet, and then, feature extraction is performed. Finally, the multigrade classification of breast cancer is executed utilizing SqueezeNet tuned by FDBO. The efficiency of Squeeze_FDBO is validated using various performance measures, and it is found to record an accuracy of 0.919, sensitivity of 0.913, specificity of 0.923, Negative Predictive Value (NPV) of 0.920, and Positive Predictive Value (PPV) of 0.908, and a better routing performance with energy of 0.405 J, distance of 6.901 m, and delay of 0.650mS.