Machine learning algorithms combined with IoT technology for continuous heart disease prediction face challenges in processing noisy sensor data compared to standard datasets.
This review highlights the challenges and potential frameworks for using machine learning and IoT technology for continuous ECG monitoring and heart disease prediction.
The major disease caused by human death nowadays is heart disease, due it happens suddenly and without significant symptoms, leads patient to miss the best time for first aid. With the development of IoT technology combined with the healthcare industry. It is providing technical support for clinic staff to predict and monitor heart disease patients remotely. In this paper, the main goal is to review the most relevant and latest papers to find the advantages and disadvantages and gaps in this area. Furthermore, compare the different proposed method's performance and present the best framework for heart disease continuous prediction and monitoring. Many researchers have been already providing the use of different types of machine learning algorithms to predict and diagnose heart disease. However, most of the previous researchers use the data collected from the dataset. As well know, to process the data collected from IoT sensors is harder than data collected from the dataset, because it may contain more noise and missing values in IoT sensor collected data. Dealing with those issues is the main challenge in the whole prediction system. Therefore, in this paper, we expect to reduce the research gap to find the best way to continuously monitoring and predicting patient ECG signals collected from IoT sensor devices in the meantime achieved acceptable prediction accuracy.
He et al. (Wed,) conducted a review in Heart disease. Machine learning based on IoT technology was evaluated. Machine learning algorithms combined with IoT technology for continuous heart disease prediction face challenges in processing noisy sensor data compared to standard datasets.