An IoMT Enabled Iterative Artificial Bee Colony Approach Using Federated Learning for Detection of Heart Disease

Several fatalities occur due to heart diseases across the world. The Internet of Medical Things (IoMT) has recently gained considerable popularity in diagnosing and detecting diseases. There have been excellent results relative to the early detection of diseases due to IoMT approaches. The data collected at run time and processed in early stages enables healthcare to save human lives. There are sensor-based nodes that support the data collection and make it easy to predict diseases. However, the private data of the patient collected by these devices paves a severe concern about an individual's privacy. The information flow in the sensors is continuous, and at times, it is bulky. The bulk of data streaming across the network nodes may prove fatal to the privacy of the individual candidates. Several issues regarding the data processed using machine learning and artificial intelligence-based heuristics are raised. In this study, we propose an Iterative Artificial Bee Colony Algorithm (I-ABC) using Federated Learning for the optimized results to reduce the secrecy and privacy of the data collected over the nodes. For the prediction of heart diseases, this algorithm will be constructive in identifying the critical values of observations from the pool of information flowing across the data nodes or sensors. The accuracy of the disease diagnosis, error reduction, and data efficiency are expected to improve with our iterative optimization technique. The result from the proposed federated learning is more profound than the various state-of-the-art methods deployed in this context.