Inspiration:
When Dr Fadi Nasrallah presented the problem statement on how retinal scanning has become a major problem for ophthalmologists during tele consulting, and how retinal scanning is a vital aspect of any eye appointment, i thought there had to be an affordable and feasible way to solve this problem.The first thing that came to my mind was my old kaleidoscope which hardly costed me 4 dollars.Thats when i thought how my grandad's magnifying glass can be integrated to a kaleidoscope kind of set up which could then be attached to my phone camera for magnified and focused retinal pictures.
Also, for the software component of the project, i had read an article few months back where a person had clicked an 81Mp resolution image of the moon using his 12Mp camera.He did this by taking 50,000 shots of the moon and superimposing them and processing using masks and filters.
What it does
My prototype design is an attachment to either the rear camera of a smart phone or the webcam of the laptop.The patient is required to peek inside it allowing the camera to take picture of the retina in the form of a video.
The software component tells the patients the length required to adjust for proper focal length.After the images are captured, it is processed with by masking or applying different types of filters.It sends the processed clear image to the ophthalmologist.
How we built it
After sketching the prototype design,i tried to convey my imagination using a roll of paper which could increase or decrease in length as and when required.
I wrote a matlab code for increasing the contrast and detecting edges in a retina image and finding the histogram equalized image.
Challenges I ran into
One of the major challenges i ran into was that i formed a team with people from different countries on discord but soon after the pitch, they became unavailable and i had to do the whole project by myself.
Accomplishments that I am proud of
I was successfully do the image processing of a retinal image and found the result to be much clearer and significant improvement of the input image
What I learned:
Not to give up even if i had to do to the whole project by myself.
I also learned certain portions of image processing from the signal processing workshop
What's next for KaleidoFundus
Product manufacturing and implementation
Log in or sign up for Devpost to join the conversation.