The purpose is to enhance clinical diagnosis by collecting data in an efficient and reliable manner. Medical image analysis has become one of the top research and development targets as a consequence of recent discoveries in biomedical engineering. As the number of data volumes created by various medical imaging modalities develops in lockstep with the usage of medical imaging in clinical practise, data compression is necessary for transmission, storage, and administration of digital medical image data sets. One of the reasons for this advancement is the use of machine learning algorithms for medical image processing. Deep learning techniques, in which a neural network learns characteristics automatically, have proven to be incredibly effective as a machine learning tool. In contrast, techniques that depend on traditional handmade parts are less effective. A wireless sensor network may be used to build a rudimentary healthcare system that collects patient data and improves medical convenience in an emergency. The WSN architecture had to integrate sensor nodes that used less power and resources at a low cost, which led to Raspberry Pi-based WSN nodes. Medical image compression methods created utilising deep learning techniques such as autoencoders and restricted Boltzmann machines (RBM), as well as the installation of WSN sensors nodes using Raspberry Pi and the MQTT IoT protocol for secure picture transfer, fall into two areas. The results are evaluated using industry-standard performance parameters like as PSNR, and an RTL implementation of the idea is also shown.