Inspiration
Veterinarians are dedicated professionals who work tirelessly to ensure the well-being of pets at home and animals in the field. However, they face significant burnout due to the manual admin tasks that come with the job. Despite increased demand for veterinary care in the U.S. and abroad, there are significant concerns about the long-term supply of veterinary professionals. Financial admins must perform slow, manual, and tedious processing of billing codes and pricing for procedures, which due to a shortage (1 admin per 6-7 vets), further contributes to bottlenecks in hospitals and clinics. Recognizing these challenges and the lack of existing solutions that meet quality standards, our team set out to develop an AI-powered application to streamline veterinary admin processes and reduce burnout, thereby giving vets more time to see additional patients and provide a higher quality of care.
What it does
Purr AI is an innovative mobile application designed specifically for veterinarians, both in clinics and in the field. The app offers the following core functionalities, which are unique in the market:
Voice-to-text transcription: The app efficiently and accurately converts the vet's audio notes (spoken words) into text during appointments or while working in the field where desktop applications are not readily accessible. Note: this was built 3 weeks before the hackathon.
Automatic summarization: Utilizing advanced Large Language Models (LLM) assisted by Natural Language Processing and Understanding (NLP/NLU) techniques, Purr AI identifies and highlights the most important points of the transcribed text, allowing veterinarians to quickly review key information.
Billing code generation: The app automatically assigns relevant billing codes based on the transcribed text's keywords, streamlining the invoicing process and filling a gap in the current market.
Predictive pricing: Leveraging a comprehensive database and sophisticated machine learning algorithms, Purr AI provides real-time pricing estimates for various procedures, reducing the burden on financial admins and speeding up the process for customers.
How we built it
Purr AI is primarily built for iOS, using React Native for the app's frontend and Firebase for the middleware and backend. We integrated cutting-edge AI and Large Language Models for voice-to-text transcription, text summarization, and predictive pricing. We also utilized a combination ChatGPT and Github Copilot to build the initial frontend scaffolding, which was around 25% of the entire codebase.
The user-friendly UI is designed with the busy veterinarian in mind. It features a minimalistic, intuitive interface that focuses on providing the most relevant information at a glance and was inspired by successful health-tech product leaders such as Ro Health. Eventually, it will be paired with a web app where vets can seamlessly export medical records to their EHR system of choice.
We also prompted ChatGPT to act as one of the worst Product Designers imaginable and provide recommendations for how to design Purr AI (as you might have guessed, we did the opposite of what it suggested…).
Challenges we ran into
We faced challenges in ensuring the accuracy and reliability of the AI models for automatic summarization and generative billing codes + predictive pricing. Developing a comprehensive database for billing codes and pricing was also a significant challenge, as this data is not standardized in the U.S. and exists in multiple places. Additionally, mapping the backend data and abstracting the app's data, triggers, processes, and functions proved to be a complex task for our team.
Accomplishments that we're proud of
We are proud to have developed Purr AI, the first veterinary app that leverages AI and LLM functionality to address the pain points of veterinarians and financial admins. Additionally, we’ve received PAWsitive (pun absolutely intended) feedback from user testing that’s been done to confirm our hypothesis that this is a problem that needs to be solved. By offering innovative features, such as billing code generation and predictive pricing, our app helps to reduce burnout and optimize veterinary practice, allowing vets to spend more time doing what they do best: saving lives.
What we learned
Our team gained valuable insights into the world of veterinary care and the challenges faced by professionals in the field. Veterinary software is an underserved niche with a lot of problems that can be solved with technology, which brings value to vets, customers, and hospital systems. This is a rare opportunity that doesn’t result in the unfortunately common zero sum game of healthcare problems and solutions: as vets benefit from the software, customers receive higher quality care and hospitals are able to process a higher volume of patients.
We also learned how to effectively integrate multiple AI and LLMs into a single, cohesive application that provides a clean end-to-end experience, and tackle the technical complexities of mapping the complex backend data.
What's next for Purr AI: Revolutionizing Veterinary Care with AI Technology
We plan to further refine and expand Purr AI's capabilities by incorporating additional functionalities, such as in-app billing (for vets in the field), secure patient data storage, and applications for Android and web. We also aim to develop Purr AI in other languages such as Spanish and Japanese. Additionally, we are exploring solutions that allow vets to bring in test results and vitals from other third party apps directly into Purr AI, which would then be integrated directly into their EHR of choice. We had a lot of fun working on Purr AI at the hackathon and we hope you enjoy using our app!
Built With
- cloudfunctions
- firebase
- firebasestorage
- gpt3.5
- javascript
- node.js
- openaiapi
- reactnative
- realtimedatabase
- serverless
- whisperapi
Log in or sign up for Devpost to join the conversation.