Underserved communities are disproportionally affected by illnesses such as heart disease, hypertension, diabetes, etc... Often due to the lack of access to quality care, they are often left on their own to manage their own medical conditions and make lifestyle changes such as what they eat. Many struggle to make the changes or lack knowledge of the effect their diet has on their health. Often times because of this their health worsens. Without help changing your diet proves to be challenging and time-consuming trying to figure out what you can eat to maintain your health and how to incorporate that diet into day-to-day deals.
What it does
Umi allows patients to make wiser choices about what they eat based on their medical conditions or dietary restrictions. For example, a patient that has been diagnosed with type 2 diabetes can find meal ideas that don't contain a lot of carbohydrates or Someone diagnosed with high blood pressure can find recipes containing low sodium. Umi doesn't only allow for recipe discovery it can help you create Meal Plans and even generate a shopping list for you to take on the next trip to the grocery market. When you search for recipes, you have the option of saving them to your meal plan for the week.
Although not original, I believe that an accessible utility like this would prove to be beneficial to communities lacking access to quality healthcare. UMI can make it possible
How we built it
Umi is built on top of the Healthbot provided by ** Azure bot service . We use **Azure logic apps to integrate with Third Party APIs such as EDAMAM and Spoonacular to find recipes that keep in mind a user's medical conditions, allergies, and diet.
Challenges we ran into
- Not having enough time and being the sole developer working on this project
- Time wasted trying to debug Actions and logic app workflows
- EDAMAM'S API only has a chronic illness filter for enterprise consumers of the API, so that called for more research on common illnesses and recommended diets for those and mapping it the filters Edamams API already has for example for a user with diabetes UMI would search using the low-carb and sugar-conscious parameters in the request. ## Accomplishments that we're proud of Finished MVP (Hopefully) ## What we learned Azure ( Bot service, functions, SQL Server, Logic Apps) ## What's next for UMI Id like to flesh out the meal plan and meal finder scenarios to handle invalid input better, better recipe suggestions. Also I'd like to create another scenario where users can enter the foods they do eat and the bot will then rate your diet based on how healthy it is and your own health condition and suggest alternatives meals ~~