Because of the incomprehensive to specialties
View the demo »
- It is very difficult for underserved communities to access the right healthcare provider with the specialties for their symptoms. So we try to design the journey to bridge the gap between them.
What it does
- Users can pick the tags that describe their symptoms well, our service will help them convert into corresponding healthcare specialties, and display the candidates of healthcare providers which meet their needs.
How we built it
- Based on the requirements above, we decide to build a website with a human-friendly interface as the implementation. We simply divide the service into frontend and backend, the former includes SDK integration and UI presentation, and the latter contains the conversion system and storagement.
- The project is driven by the Scrumban cycle, we maintain a public bulletin board and a knowledge base using HackMD throughout the development, it allows us to converge ideas quickly and control the schedule and resources.
- We finally deploy the service on Microsoft Azure via CI CD pipeline, maintaining the equipment by ourselves is too much trouble for us, the free quota of public cloud is a great choice for the hackathon.
Challenges we ran into
- Our developing resources are really limits, the server on Azure for free account only has 1 CPU and 1GB RAM, so we need to optimize the usage efficiency.
- The Google Maps Place API we used has a limit on the number of times, caching the information helps us reduce the usage amount and the response latency.
- The fragmented description in the documentation confuses developers, we figure out the pattern of the content through try-and-error.
Accomplishments that we're proud of
- Our service provides a relatively low barrier for the underserved community, we believe that an easy-to-use process is a key to problem-solving.
- We give a full play to the advantages of the SDK, and our conversion system builds a fluent journey to the healthcare providers.
- Our whole system is fully cloud-native, we deploy our services on the Microsoft Azure platform and can scale out effortless.
What we learned
- The hackathon is a rare opportunity for us to practice the user-oriented service design, we collect a lot of statistical data to portrait the persona of our target audience at the very beginning.
- Cooperate with each other remotely at a fast pace is also a challenge, but we found determining the specification and action items on the first day of each development cycle helps us deliver features efficiently.
What's next for hcpairing
- The de-identified records of the user searching could be stored persistently in the database, maybe it can be helpful to the NGOs in planning strategies.
- Pin the critical information on the top of the searching result also worth a try, a short version of policies can directly fill up the information gap.