In the United States alone roughly 37% of dogs are overweight and 19% are obese. This equates to more than 50% and research has found that these numbers will continue to grow. More often than not dogs are overfed or given poor exercise leading to severe health issues and decreased life expectancy. When we talked to veterinary professionals, they told us how they usually have to calculate the lean weight of the owner’s dog to update the food amount given to their dog on a daily basis. The problem is dog owners usually only go to checkups once a year. That means that the food amount will eventually be invalid, and the dog may be eating too little or too much based on its lean weight progress over time.
What it does
Our app takes in user input of a dog's characteristics and information to determine how much food and exercise the dog should be receiving on a daily basis. What’s unique about our app is that we have an API that will progressively update the age of the dog, and we have our own python machine learning predictive model using Keras that will predict what the lean weight of the dog will be over time since we already know how much the dog is eating every day. The predictive model is graphed on the nutrition page, and you can see key weight milestones for the dog such as when it’s 1 year old, 2 years old, wtc. We also have additional features such as a Vaccine Log that will help the user keep track of all past vaccines and a Behavior Log that will help owners track significant behavioral changes which could ultimately help veterinarians diagnose diseases and problems sooner.
How we built it
We used the Flutter SDK by Google to create an app that can run both on the app store and the play store. The primary language we used was dart, and we worked as a team and split up different pages of the app, and once we were done with our individual work, we merged everything using GitHub. Our key emphasis throughout the app was to have a simple but functional user experience. Most other pet tracking apps only track physical activities or calorie count, but require the user to enter in the calorie amount and exercise time every time the dog eats or wants to go out. We knew this can be disheartening for a consumer to consistently use an app, so our main focus was to have one setup process where you set up your dog’s information when you first download the app, then the user will never have to edit anything again. We also chose a lighter tone for the UI of the app to make it seem more simplistic to the user.
Challenges we ran into
When submitting we ran into technical difficulties since our Zoom calls weren’t working. We also had a lot of challenges with the Flutter software. We also experienced a lot of problems with the animations throughout the app. We wanted to make the user experience as appealing as possible, so our decision to spend a significant amount of time on it led to numerous animation bugs and rendering problems. Issues with implementing the API and other UI navigation issues constantly popped up as well but we worked through it together, and we are extremely proud of the product we produced.
Accomplishments that we're proud of
We are most proud of the nutrition page because there is currently no app that can predict the lean weight of a dog over time, and determine the exact amount of food the dog should be eating. The model required the implementation of an API that could determine time zones, and update the age of the dog over time. This implementation in itself was difficult for us, and we were very proud to see everything work in cohesion. We were also proud of the animations throughout the app and transitions between screens because these ultimately made our app look more professional and appealing for us to use.
What we learned
We learned a lot about how to make an app, and specifically how to implement an API and python back end. This was the first time we had built an app through the Flutter software, and we were excited to create an app which would work for both IOS and Android. We spent quite some time working through the small errors so it was very nice to solidify the basics. We did have experience with the android studio so it made it easier for us to build this app as a whole, but we learned a lot about Flutter and Dart.
What's next for PetCare Tracker
So currently we are focusing on creating a fully functional and viable prototype as there are many features and tweaks we would like to make. Once we do that we plan on running a few Beta trials with Pet Owners and their Vets. At that stage, the revenue would be generated through ads. If our trials prove to be successful we will file for a proprietary license and copy-right the code. We’ve read the USPTO criteria and we believe our app would be eligible for a patent as well since with combination of features, especially the predictive model. For our third stage, we would also generate revenue through a licensing fee if we can work with pet clinics.