Inspired by the Treehacks Health Challenge and our interest in the intersection of software and healthcare, we wanted to create something that could have a positive impact in the health industry. We took on the challenge inspired by Dr. Peter Karth’s work and the voiced need for a mobile application that can increase the accuracy of digital vision test.

What it does

Unlike current vision tests that use manual input to record answers and do not guarantee that a user is at the proper distance throughout the exam, Eye Vision Test is an iOS application that uses an open source speech recognition API, bluetooth and ibeacon technologies that allows user to verbally register their answers and tracks how close or far away the user is from the device to make sure they are at a fixed proper distance through out the exam.

Eye Vision Test connects two devices together, an iPad and an iPhone through bluetooth. The iPad serves as the exam hub, which displays an eye chart. The iPhone serves as a remote that helps the user communicate with the exam hub. The connectivity of the devices also helps detect the distance between the user and the exam hub.

The iPad displays one letter a time, it waits for the users' verbal response of what letter they see. The size of the letter presented is determined by an eye chart algorithm created in collaboration with Dr. Peter Karth that analyzes the users response and it also calculates an accurate vision score.

Challenges we ran into

We initially wanted to use Alexa’s Voice Service for the speech recognition portion of our app, but due to a server issue on Amazons end we were not successful in integrating the service. We also turned to IBM Watson for their speech to text and text to speech API’s, but their libraries were coded in Swift and our language of choice for this application was Objective-C. Due to this, we resorted to using an open source Speech Recognition API.

Our initial application architecture was going to use the bluetooth connectivity and location services to detect the distance between the exam hub and the user, but we came to the realization that using ibeacon technology over bluetooth would give more accurate distance detecting in smaller distances. Due to time constraints this feature was not fully implemented.


We successfully paired two devices through bluetooth and we also managed to integrate the open source speech recognition framework at the last minutes and were able to see two devices communicate with one another through speech.

What we learned

We learned the power of voice recognition and how it can enhance the experience of the user while providing more accurate results for digital vision tests. We also learned how to successfully pair two devices automatically. Lastly, we learn more about how eye examinations are evaluated.

What's next for EyeVisionTest

Short term goals

-Finish implementing the ibeacon technology for distance detection.
-Try to integrate Alexa Voice Services, IBM Watson or Microsofts Oxford voice services for smarter voice recognition abilities.
-Improve the UI/UX of the application.

Long term goals

The main purpose of this application is to provider users who have chronic vision problem and need to monitor their vision on a regular basis a digital vision test. This digital vision test should provide accurate vision results and could be taken from the comfort of their own home instead of the requiring the patient to visit the optometrist office. It should also notice the optometrist of any alerting vision changes. With this in mind, our long term goals for this application include:

-Incorporate the notion of user accounts
-Record users vision results through time
-Provide email communication between patient and doctor

Built With

Share this project: