Real-time Obesity and Heart Disease Monitoring with IoT and Deep Learning

Abstract

The accelerating spread of junk food consumption is progressively putting the younger generations at risk of obesity. The concept of health prioritization is deeply rooted in the development of nurturing and general welfare in society. The job has further become more complex by the glaring statistics of people across the globe who are fighting obesity. The World Health Organization (WHO) puts a staggering estimate that approximately 1 billion people across the globe suffer from obesity-related issues, a statistic heavy in terms of impact primarily because obesity greatly affects one's health but also holds fatal risks. Diagnosing the early stages of obesity can just be simple and prevent one from being obese; however, several obstacles stand in the way of its early diagnosis. Symptoms or signs that one is in the trajectory of suffering from obesity generally go unnoticed by the individual, and the availability of specialists to diagnose such risks persistently on time is not always possible. In this regard, the current paper shall showcase an application hitherto unknown using cloud-based diagnosis, in collaboration with machine learning through Secure Access Internet of Things to predict obesity-related risks without having to depend on glaring physical parameter indicators—in this regard, redefining how such information could be sourced even without the need of regularly visiting health professional people. To ensure the best possible health predictability from this machine learning, we gathered and plotted the datasets consisting of obese and overweight people. Other than machine learning, we use a Secure Access IoT sensor network. This allows for real-time monitoring and diagnosis, making early detection achievable. Consequently, this system paves the way for mitigating and substantially reducing the complications associated with obesity.

Topics & Concepts