We are the leading health care app development company in US, UK and Canada. The client wanted to develop a smartphone app that estimates newborn jaundice levels for a Norway Semi-Government project. We developed a user-friendly healthcare mobile app that estimates newborn jaundice level to ensures that anyone can use the app after seeing just a short video tutorial. According to the statistics every year 114.000 newborn children die from jaundice. Although jaundice rarely causes death or brain damage in high-income countries, the condition is still a large problem in low and middle-income countries. This is due to the high cost of the medical devices used to accurately diagnose the condition today. We are awarded the best health care app development company by Clutch. View our PORTFOLIO and CASE STUDIES to know more about latest project related to healthcare and fitness.
The main challenge of building this app was to develop a mobile app for both iOS and Android. The challenge lies in the fact that the app built in both the platforms has to take the pictures of the newborns, provide the required data and estimate the Bilirubin value.
We discuss to ensure that we have the exact idea of what is required
There's regular interaction with the client to ensure things are on track
Begins according to the needs of our client
The final output will be a perfect match to our clients requirement
Initially we started developing the application which will let the Doctors and Nurses who use the app to take the pictures of the newborns and send it to the server to calculate the Bilirubin content. But due to different camera quality and light effects, the accuracy of the calculation lacks, which is a serious issue as far as the medical field is concerned.
So we came up with a solution to use a reference card for the calculation. The reference card is a multi-colored card that is the size of a credit card, which the user can put on the newborn’s body and based on these references the photo will be sent to the server. As accuracy is the main concern here, we added some additional care to it.
All the six images along with some information of the newborn are sent to the server to calculate the Bilirubin value. The server will refer the images based on the color codes on the card and the newborn’s body color, then estimate the value of Bilirubin with the medical calculation and send back the result to the app.
We used Flutter as the development platform to develop this application. As Flutter was a new development platform when we started the project, there were no efficient libraries to manage the image processing, auto capturing and card detection. So we needed to take an OpenCV_flutter camera plugin and do our own modifications to make the app manage the card recognition, auto capturing and image processing. This OpenCV library is the major risk we had while at the stage of development. We managed to solve the glitches in the library and make the app working.
The next risk we faced is the calculation process on the backend system. Initially the system took more than 5 minutes to process the image and send the results back on to the app. So we managed to use the same OpenCV library to process the image on the server side as well. As we were using the same library to capture and process the images, the calculation speed increased and the calculation time reduced to less than 60 seconds.
Flutter, OpenCV, Python,
Xcode and Android Studio
Atlassian Confluence, Gitlab.
When this app is considered, the purpose and usage of the app is for those who work in the medical field. So our approach is to make a simple and easily workable UI for all kinds of users. So we created a UI with a simple camera and an instruction button. The instruction is a video tutorial which the user can watch any number of times if they want to refer on how the app works and how they need to make it work. When it comes to the camera option, for greater UI experience, an auto capture and auto flash option is provided for capturing the images which will let the app sends the best images to the server with great quality.
The project received grants from and funding from European Union’s Horizon 2020 research and innovation program under grant agreement no: 854926