Abstract
Saudi Arabia is lagging behind developed countries in devising specific real projects, roadmaps, and policies for the Internet of Things (IoT) and big data adoption despite having a vision for providing the best-quality health care services to its citizens. As a result, Saudi Arabia is going to host an event for the third time, in 2020, promoting the widescale adoption of the IoT. While a nationwide study has identified the risk that many participants were previously undiagnosed for hypertension and other chronic diseases in Saudi Arabia, the application of the IoT and big data technologies could be very useful in minimizing such risks by predicting chronic disease earlier, and on a large scale.
A framework that consists of four modules, (1) data collection, (2) data storage, (3) Hadoop/Spark cluster, and (4) Google Cloud, was developed in which decision tree and support vector machine (SVM) techniques were used for predicting hypertension. There were 140 participants in total and 20% of participants were used for training the model.
The results show that age and diabetes play a very significant part in diagnosing hypertension in older people. Also, it was found that the possibility of hypertension because of smoking is less than that of diabetes, and older people should have a lower intake of salty food. Moreover, it was found that SVM techniques yielded better results than C4.5 in our study.
Although it was found that the algorithms examined in this study can be used for disease prediction, the ability to classify and predict disease is not yet sufficiently satisfactory. To achieve this, more training data and a longer duration are required. Finally, by supporting such study for developing custom-made smart wristbands, custom-made smart clothing, and custom-made smart homes that can predict and detect a wide range of chronic diseases, the Saudi government can achieve its health-related goals of Vision 2030.