Guest Post by Josh Leibstein
Student and Imagine Cup finalist
When I developed the computer-aided detection programme to identify tuberculosis at an early stage, I never imagined it would lead to winning first prize at Microsoft’s South Africa leg of the Imagine Cup. The competition was tough as there were many great projects – all of them meeting this year’s criteria, which was ‘Solving the World’s Toughest Problems’!
It all started a couple of years ago when I overheard my fellow University of Johannesburg students discuss their Imagine Cup entries. It intrigued me. Immediately, I was inspired by the extent of their creativity and that their projects actually had the potential to address real-life issues. And I thought – hey, I can do that too!
Under the guidance of my honours project mentor, Mr Duncan Coulter, I started working on an image processing application. The great results of the project turned out to defy even my biggest expectations, as I realised it could be applied to help solve a serious problem. Mr Coulter was instrumental from initial design through to the final implementation. His general guidance, feature suggestions and design ideas were incredible.
The project started out as an application that statistically analyses the texture properties of arbitrary images such as tiles, clothing or paper. Using that data, I found areas that are similar to other given samples. Once I was able to identify and analyse those types of textures, I observed the merits of applying it to more complex textures, such as those that need to be analysed in the early detection of tuberculosis.
I realised the imaging system would be perfect in assisting a radiologist in identifying areas that have a high probability of containing tuberculosis structures as result of their texture properties.
Tuberculosis is an endemic disease affecting South Africa on so many levels, socially and economically, and it made sense to me to expand my thinking to try and address this issue. The results so far have been quite promising. As it turned out my project fit perfectly into the theme of the Imagine Cup.
What’s more, I get to represent my country and pick the innovative brains of like-minded fellow contestants at the Imagine Cup in Sydney later this year.
Innovation in South Africa is at such a promising stage, especially among the youth. We have many talented up-and-coming developers. Competitions like Imagine Cup gives them an awesome opportunity to produce great ideas for the future – possibly even "the next big thing". The projects showcased at the last national Imagine Cup finals demonstrated this.
With the right mindset our students can compete with anyone in the world. But we have to overcome this notion of trying to imitate ideas from the USA and other countries. We as South Africans have our own style and we use technologies that give us a unique edge. Individuals should be encouraged by business and government alike to create cutting edge solutions, rather than take the safe option.
My view is that as our students move into industry, this attitude will hopefully filter through to the public and private sectors, making South Africa a top competitor globally.
As for my own dreams – I will continue development of the project as part of my master's thesis and hopefully the system can undergo field trials in the not too distant future. The potential for this research to assist so many people is a great motivator. If the system can help stem the tide of a rampant TB epidemic in our country, it will all be worthwhile.