How likely will a person develop a heart disease condition within the next ten years? Our following BOTX tutorial shows how to create a machine learning model using neural networks, DCS (our integrated data platform), and an online assistant to predict this very question based on 15 parameters.
This blog post and video are part one of the tutorials focusing on the machine learning model and DCS setup. Play the tutorial in the YouTube below.
Around 17.5 million people die each year from cardiovascular diseases (CVDs), an estimated 31% of all deaths worldwide. This statistic is expected to grow to more than 23.6 million by 2030. Of these deaths (17.5 million), 7.4 million are due to coronary heart disease, and 6.7 million are stroke. Epidemiologic studies have played an important role in elucidating the factors that predispose to CVD and highlighting opportunities for prevention. Most CVDs can be prevented by addressing behavioral risk factors such as tobacco use, unhealthy diet and obesity, physical inactivity, and harmful use of alcohol.
Our understanding of the above critical facts about heart disease was due mainly to research known as the Framingham Heart Study (FHS), the most influential investigation in the history of modern medicine. It is a long-term, ongoing cardiovascular study on residents of Framingham, Massachusetts, USA. The study began in 1948 with 5209 adult subjects from Framingham and is now on its third generation.
Download the data here. You can find the data from which our dataset was derived on kaggle.com. More information about the FHS is here.
We can see that BOTX is a powerful tool that can be used by non-engineers and achieve excellent results. In the second part of this tutorial, we will discover how to create an online assistant and let the two robots talk to integrate the neural network into the assistant.