Trillium is an iOS application allowing users to recognize the results of COVID-19 antibody tests contracting.
Trillium
-
Platform
- iOS
-
Devices
- iPhone
-
Tech stack
- Swift, .NET Core, Amazon RDS, OpenCV, Angular Smartadmin
-
Industry
- Array
-
Scope of work
-
iOS client development, the back-end and front-end development, software design and architecture, deployment, quality assurance
The challenge
The client created a highly-efficient custom device for COVID-19 tests that provides results in 15 minutes. However, the users cannot interpret the results of these tests without a precise test analysis tool. That’s why client asked us to built an app based on the image recognition of tests for the most rapid interpretation of COVID-19 test results. The app uses complex math algorithms to interpret test results and provides users with instant and precise results regardless of the camera tilt, camera shake, etc.
The solution
Our developers built an iOS application that provides users with step-by-step instructions on how to take the blood samples, scans the results in the test window, saves them to a database, and then shows the results right in-app.
The client required accurate recognition tool, considering different user conditions (e.g. lighting, phone shaking). So, we found an approach to these situations using the built-in accelerometer to give the most accurate results.
The approach
Our experts leveraged math algorithms to scan the tests and explain them. For the stable operation of the iOS app, we used the proven Swift technology. Also, our professional team implemented the .NET Core web API, Amazon RDS, and Angular Smartadmin template for the Admin panel. We also provided the app with the privacy of the patient’s data ensuring HIPAA compliance.
What is most interesting in this project is the solution to stabilize the reading of the COVID-19 tests. Using Vision, we determine whether there is a QR code in the photo to show the user feedback on what needs to be pointed at the cassette. With the help of the OpenCV library, the tool recognizes the area of the test. For the resulting image, it changes the colormap with OpenCV as well. This algorithm helps us deal with shadows, poor lighting, and low-quality tests.
End results
As a result of our effort, the client received an efficient image recognition app for the most precise and rapid decoding of COVID-19 test results.
ABCloudz has all the required expertise for building the most efficient image recognition apps not only for Healthcare industry. With our custom solutions, software will capture and decode texts or images regardless of illumination, definition, tilt, or any other factors.
Talk to our consultants and leave all the project’s puzzles to us!
