Internet of medical things (IOMT) is a revolutionary newly emerging technology that is capable to increase accuracy, reliability, and productivity in the healthcare sector. A smart 24/7 healthcare system that can detect health conditions daily is a significant problem, particularly in poor countries with a shortage of high-quality hospitals and medical professionals in remote places. In this paper, we propose an Intelligent e-healthcare model for the early and precise identification of breast cancer, called IEMBC as an IOMT application. The proposed IEMBC model intends to be a low-cost solution in the healthcare sector to serve people in remote areas. For the accurate extraction of infected regions, a hybrid cross-entropy thresholding technique is adopted. The most creative aspect is the parallel boosting approach, which was created and deployed to improve the performance of the suggested model in cloud-based e-healthcare services. The proposed IEMBC model employs a deep learning approach for identifying breast cancer to obtain higher accuracy. The suggested approach's performance was evaluated using a variety of real-world breast cancer datasets, including CBIS-DDSM and INBreast. Experimental results highlight the efficacy of the IEMBC model, including (1) achieving up to 93% segmentation accuracy with the proposed IEMBC-HGG unit, (2) reducing processing time by 66.87% with the parallel boosting approach (IEMBC-PB), and (3) achieving an accuracy of 86.32% in breast cancer classification using the IEMBC-DC unit, showcasing its robustness and performance benefits.
