Inspiration
Many diseases such as diabetes, heart disease and lung cancer disproportionately affect the black community. Many of these diseases are also either caused by poor dietary habits or can be prevented by healthy eating habits. We thus wanted to create an easy way for individuals to monitor the nutritional content of the food they consume and adjust their diets whenever necessary.
What it does
Individuals take pictures of their meals, and upload it to the Health Watch website. We then compile weekly nutritional reports, which include recommendations on how individuals can improve their eating habits, such as by reducing their sugar intake. The recommendations we make will adjust recommendations based on factors such as age, race and fitness level of the individual.
How I built it
The project was divided into 3 categories. The front-end of the website, processing images using computer vision and using those data output from those processes to come up with a semi-accurate nutritional break-down of the users daily food consumption. The first part would have been done using CSS, PHP, and HTML. We were also planning on using the Google vision library in Python to process our images; we however were unable to learn how to properly use the library within such a short amount of time. We intended to take the output of the Google cloud vision and extract a series of keywords from the output, such as steak, chicken breast etc. Furthermore, we would use the colour breakdown of the images to calculate proportions for each food. Finally, we would use the data we had to determine information such as whether the individual was consuming too much sugar and send them the appropriate recommendations.
Challenges I ran into
A lack of expertise on Google Cloud API, and general implementation of the software. We also wasted a lot of time trying to figure out how to use the Vision API.
Accomplishments that I'm proud of
We're proud of our idea and the potential impact it can have on the black community. We are also proud of what we were able to accomplish despite facing a significant amount of challenges.
What I learned
We learned a significant amount of things within the space of about 6 hours such as the basics of the Google Cloud platform and Vision API.
What's next for Health Watch
Eventually, we want Health Watch to send these weekly reports to the personal physicians of users, who can further make recommendations and also be able to catch conditions early.
Log in or sign up for Devpost to join the conversation.