SmartKC

Smartphone-based Corneal Topographer for Keratoconus Detection. Keratoconus is a severe eye disease affecting the cornea causing it to become thin and develop a conical bulge. If not treated in time, keratoconus can lead to near-complete blindness.

The diagnosis of keratoconus requires sophisticated ophthalmic corneal topographers which are non-portable and very expensive. We have developed SmartKC, a low-cost, smartphone-based keratoconus diagnosis system comprising of a 3D-printed placido’s disc attachment, a LED light strip, and a smartphone app to capture the reflection of the placido rings on the cornea.

Collaborative project with a wonderful team from Microsoft Research India.

Published Research Paper -

Additional Links

https://www.microsoft.com/en-us/research/project/smartkc-a-smartphone-based-corneal-topographer/

https://www.youtube.com/watch?v=rp4uyzf6e2Q&ab_channel=SiddharthaGairola

Back to Top_SmrtKC

Initial Set up

The initial proof of concept for the idea was developed with a circular disk attached to a rudimentary and temporary setup. It was using this models that the concept was validated and the journey to the actual project began.

Testing and getting to know the Keratron machine

Keratron has been the Corneal Topography standard for a number of years now and was natural that we had to get to know the machine and its functionality before we design our version of it.

Ideations

The next step was to get back to the drawing board and start with the ideations. All the designs were through through and the calculations were made to make sure the concept turned into a working product.

Initial design 1

Some initial designs were modelled and 3D printed to validate the concept again. This one here was a simple model that was made that can be screwed on to the mobile holder. The concept was to have different rings of varying dimensions attached to the same body.

Initial design 2

This model here was another ring model that can be attached to the camera body. Testing with multiple rings gave us an idea on what to finalize on when moving towards computing the ring reflections in the mobile app later.

Mobile attachment design

Once the ring patterns were finalized the ways to attach that to a mobile using various holders were then explored. It had to be simple to attach and remove as well as easily replicable. The ring structure was also designed to make sure it can be detached from the clip design if needed and did give structural strength to the ring assembly.

Different Outer structure designs

Even though the rings dimensions were finalized, we faced a challenge of making sure the rings were protected from mishandling as well as enclosing them made sure there were less scattering of light there by getting good clarity reflections of the rings.

Outer body designs

A lot of factors were considered for the outer body structure shape main among which were the reflection of light by the rings at the top, the sturdiness of the body etc.

Final structure

After a number of iterations and 3D models we had finalized on this design as it had very less interference for the light to pass through the rings there by giving good ring reflections.

Light attachment

The led lights were sourced and and was placed circularly in the body of the clip onto which the rings were attached. Addition of the light gave great quality reflections for the mobile
algorithms to process.

Final assembly

The whole attachment can be clipped on to the mobile with the mobile in line with the main camera hole and the light can be attached to the charging port in the mobile to light it up. The app then gets the image when we look through it and process the image to see if there is any anomaly with the users eye.

In action

A smartphone app to capture the reflection of the placido rings on the cornea. An image processing pipeline analyzes the corneal image and uses the smartphone’s camera parameters, the placido rings’ 3D location, the pixel location of the reflected placido rings and the setup’s working distance to construct the corneal surface, via the Arc-Step method and Zernike polynomials based surface fitting.

BACK TO PROJECTS

NEXT PROJECT-Case Designs