SeeFood

Slides

https://docdro.id/wrTEfc5

Inspiration

The inspiration behind this project has been an understanding that I have the ability to create scalable solutions for some of the largest problems in healthcare solution. My grandma has High blood pressure, high cholesterol and diabetes. These are conditions that can be controlled with the correct dieting and fitness routine.

What it does

SeeFood takes photos of food, feeds it into our food recognition Machine Learning model and determines the calories and macronutrients based on the classification and recommends you different types of foods based on your health profile that is created using patient preconditions derived from Humana data.

How I built it

I used Javascript, NodeJS, intersystems, Calorie Mama, Humana, RapidAPI and Edamam Nutrition in order to bridge the gap from figuring the food recognition model all the way to the recommendation engine in 14 hours.

Challenges I ran into

Not being an iOS engineer. I made the app in html, css and javascript and made it mobile responsive so that I can demonstrate how the app would work without a full native application and steps to proceed further.

Accomplishments that I'm proud of

I was able to build the whole MVP in less than a day, but the work is not done. I still struggled to create a more scalable application with all the features I wished were embedded.

What I learned

I learned a lot about HTML5 canvas and how to capture images within videos and feeding that into a data pipeline to be digested by external API's and really scope out a project requirements before starting.

What's next for SeeFood

$$$.

Built With

• azumio
• caloriemama
• edamam-nutrition
• humana
• intersystems
• javascript
• rapidapi