Have you ever wondered why cardiovascular diseases are the leading cause of death worldwide, responsible for over 17 million deaths annually? Well, we were given this opportunity while participating in the Summer <Code> Camp at Poatek. And we discovered that those deaths occur because cardiovascular diseases, also known as CVD, are silent diseases. You won’t see them coming until you are sick.
Early detection and prevention of cardiovascular diseases are crucial to reducing their impact. Unfortunately, traditional methods of detection, such as blood tests and electrocardiograms can be invasive, time-consuming, and expensive. Because of that, we questioned what we could do to reduce the mortality of CDVs. And it finally came to us: predict to prevent! That’s how the CardioCare project was born.
The CardioCare project aims to revolutionize the strategy to predict and monitor cardiovascular diseases by developing an app that directly takes the information needed from wearables. To conclude the best strategy to bring our idea to life, we discussed the matter with our supervisors and had many Knowledge Transfer sessions on Pandas, Data Analysis, and more. Finally, we concluded that this innovative project should use machine learning methods and a public dataset to predict the likelihood of the user having a cardiovascular disease.
By analyzing a person’s vital signs, such as heart rate and blood pressure, in real-time, CardioCare, with the support of wearable technologies, will provide early warning signs of potential health problems and allow individuals to take proactive measures to reduce their risk.
The development process splits into several stages, including data collection, data analysis, and algorithm development. The data collection stage is critical as it involves gathering and processing large amounts of data from individuals with and without cardiovascular diseases. This data trained the machine learning algorithms and improved the accuracy of the predictions. Our selected method is a detection method that uses the health data given by wearables to predict the cardiovascular situation of the user. After we refined our model, its accuracy went from 98% to 99%.
The CardioCare project will change the way we detect and monitor cardiovascular diseases, making early detection and prevention more accessible and affordable. The success of this project will result in a decrease in the mortality of cardiovascular diseases. By working together, the CardioCare team can create an effective and user-friendly app, providing individuals with the tools they need to take control of their cardiovascular health. Nevertheless, we hope to be able to provide all that without asking for a single piece of data, everything is automatically detected by the wearable.
Regarding our future plans, we expect to face challenges in concern of the LGPD (Lei Geral de Proteção de Dados, the Brazilian law on data protection) and the monetization of our product. We are looking forward to having our project available to the public.
This article was written by Poatek Summer <Code> Camp Interns Ana Carolina Bortolomiol Passos, Anita Sehn Brose, Rafael Garcia Garcia, Teodoro Eilert Trevisan and Victoria Shen.
