Internet of medical things and cloud enabled brain tumour diagnosis model using deep learning with kernel extreme learning machine
M. Ganesan, N. Sivakumar, M. Thirumaran, T. Vengattaraman
Abstract
Presently, internet of things (IoT) and cloud-based e-health services offer various decision support systems in the medical field. In this view, this paper introduces a new internet of medical things (IoMT) and cloud-enabled brain tumour (BT) diagnosis and classification using deep learning-based inception model with the kernel extreme learning machine (KELM), named DLIM-KELM. The proposed DLIM-KELM undergoes a series of steps namely data acquisition, pre-processing, optimal multi-level threshold-based segmentation, Inception v3-based feature extraction, and KELM-based classification. Besides, firefly (FF) algorithm is applied for the selection of optimal threshold value in Tsallis entropy-based segmentation technique. The application of Inception v3 and KELM models helps to effectively diagnose and classify the occurrence of BT from magnetic resonance imaging (MRI) images. The DLIM-KELM model is tested using the BRATS2015 dataset and it has attained maximum sensitivity of 98.45%, specificity of 98.34%, and accuracy of 98.91%.