saveYourLife

Inspiration

According to the data from CDC, the obesity prevalence in the USA has reached 42.4%, and there are no significant differences between men and women among all adults or by age group. Obesity can be associated with some serious health risks and further increase the risk of obesity-related complications. The high prevalence of obesity is a warning that needs to be paid attention to and can be addressed by a personal effort from the weight-gaining groups.

What it does

Besides recording user's personal information like weight and height to provide a BMI value indicating the obesity level, our project tracks users' daily dietary intake, fitness activity, mood scale, and sleeping time to learn the mode that how a certain user's BMI is influenced by these factors, and then by predicting BMI with the same continuing mode, gives warnings if the prediction indicates there might be the risk of obesity or other friendly healthy reminders. A chatbot notices the daily records and responses suggested proper calories intake or exercise video.

How we built it

Built in Python, Interface: React, Could Database: CockroachDB

Challenges we ran into

We took much time and effort in brainstorming our ideas since all our teammates are new to the medical field. And there is a demand for bunches of data from users to do the reliable prediction, while we could not seek out enough records due to the patients' privacy.

Accomplishments that we're proud of

We are focusing on a serious disease problem that happens a lot nationwide, and we do provide a possible solution to notice the potential patient pay early attention to the related healthy attributes to avoid obesity.

What we learned

For us new in the medical field, we learned the background information of the US prevalent diseases and some specific aspects of our topic obesity. We created our cloud database through CockroachDB, which is also new to us, but that is a good start to learn how to use the cloud DB.

What's next for saveYourLife

We are looking forward to interfacing with the actual clinic to get more detailed health data like the blood/urine test result to reveal the latent cause and possible consequence based on the more comprehensible prediction.