Submitted by Cisco Systems, Inc.
Now that the Cisco Global Problem Solver Challenge 2021 winners have been officially announced, we are excited to introduce each winning team and the story behind each innovation. The Cisco Global Problem Solver Challenge is an annual competition that awards cash prizes to early-stage tech entrepreneurs solving the world’s toughest problems. Now in its fifth year, the competition awarded its largest prize pool ever of $1 million USD to just 20 winning teams from around the world.
The intrepid Envisionit Deep AI team was awarded one of this year’s $50,000 USD Second Runner-Up Prizes. Their AI solution called RADIFY can be used as a clinical decision support tool to assist doctors in the diagnosis of diseases through medical imaging analysis.
I sat down with Envisionit Deep AI’s three founders, Dr. Jaishree Naidoo (CEO), Terence Naidu (Head of Business Development), and Andrei Migatchev (Chief Technology Officer) to discuss their revolutionary AI solution and how it came to fruition.
What problem is your technology solution trying to solve?
Jaishree: In 2010 I became the first South African trained pediatric radiologist. It’s now 2021 and I’m still one of only 20 pediatric radiologists on the African continent, where 600 million of the 1.2 billion population is under the age of 15. Our solution aims to assist with the disparity between the number of children under the age of 15 and pediatric doctors able to serve them.
I have 20 years’ experience working in public hospitals and have participated in outreach efforts throughout Africa. Through the years and through my experience I knew I was making a difference by training more pediatric radiologists, but I realized that with just human capacity alone we cannot solve some of the challenges we face. I also knew technology could be transformative so with my medical background, Andrei’s technical experience, and Terence’s business experience, we came together to develop RADIFY.
The vision behind RADIFY, our AI solution, is to democratize access to diagnostic healthcare in Africa. By using RADIFY, doctors can diagnose more images, more consistently, in less time and at a lower cost – whilst prioritizing care for people who need it most.
In just two years, Envisionit Deep AI has developed five AI products; RADIFY for Chest, RADIFY for Mammograms and RADIFY for Ultrasound. Our RADIFY for Chest model helps with the detection of 20 different pathologies through chest x-rays and has been expanded to a standalone Tuberculosis platform, as well as a COVID-19 pneumonia platform. RADIFY for Mammograms assists in the diagnosis of benign and malignant breast lesions and RADIFY for ultrasound identifies the presence of a specific pathology on the image/video frame including the location of the pathology.
Our AI solution can assist in delivering impactful and feasible medical solutions to countries where there are significant resource challenges.
How does the solution work?
Terence: What we’ve done at Envisionit Deep AI is leverage the use of machine learning and the expertise of radiologists to create RADIFY. The solutions recognize patterns in imaging to help doctors diagnose faster. It’s pattern recognition.
Jaishree: Yeah, so it’s pattern recognition of topologies, or signs of diseases. Our AI recognizes the pattern and then annotates it on the image. So, RADIFY doesn’t just tell you an x-ray has got signs of pneumonia. It indicates what the severity of the pneumonia is and where in the lungs it (or other pathologies) are localized.
We also have a tool that can triage. So, if RADIFY examines 2,000 x-rays, it will prioritize the ones with the most urgent or life-threatening diseases at the top of the list. Since doctors will typically see patients on a first come, first serve basis, it’s unlikely that doctors would know someone in the back of the queue has a more severe disease. However, as the patients x-rays are done, our solution can prioritize the ones that are most urgent on the top, so the referring doctor has instant access to it. Also, RADIFY doesn’t wait for the radiologist report, because depending on where you are, there could be delays in getting the report.
Terence: There are even times when patients come to the cities, to big hospitals to get help, only to be told to go back home and come back in a month. In that time, not only can the disease spread rapidly but they’ve also got the cost implications of additional transport to and from the hospital and taking time off work if they’re lucky to have a job. This tool just fast tracks all of that and by the time they’ve put their shirt back on, they already have an initial diagnosis and can get treated immediately.
What inspired you to develop this solution?
Terence: There comes a point in your life and in your career where you want to have significant impact. In fact, exponential impact. We have three exceptional founders who are passionate about what they do and who are driven by more. More on solving challenges and creating positive impact. Envisionit Deep AI spoke to my heart, to my entrepreneurial spirit and to the impact it’s going to have.
Andrei: I’ve been an entrepreneur for the past 11 or 12 years and I have two passions; Information Security and Information Security. I’ve done quite a bit in my career including consulting with quite a few big organizations in South Africa. I was a Certified Ethical Hacker before I moved into the biometric space and I co-founded another company with other people.
But that all just felt like not enough. I wanted to do more. I’m getting to the point where I need to leave some sort of a legacy behind me and I believe that this could be that legacy.
Terence: I think you touched on something important: we feel a lot. That’s what really drives us. That feeling of making a difference, leaving the world in a better place than when you got here. I was born nine days before the riots in Soweto and so that’s a low base. But I think we’re all driven by the fact that we want to leave the world in a better place than when we got here.
How has the global pandemic impacted your work?
Jaishree: Envisionit Deep AI launched its RADIFY AI product in February last year (2020) and at that point, the model had seen cases in an ICU setting–the specific type of disease that causes pneumonia, which is very similar to how COVID looks. At that time we were on a venture scale program with one of the big hospital groups in South Africa but unfortunately, because of the COVID pandemic, there were lots of opportunities that were previously aligned that we couldn’t leverage off because doctors had to focus on basic needs in our hospitals.
Even though we lost the opportunity with them, we said, “let’s focus on the product, and let’s create a solution for COVID” and that’s how we got to repurpose our model for COVID. During the pandemic, we were able to repurpose our chest model and train it on patients who actually had COVID-19 pneumonia and it now prioritizes cases that have a high suspicion of COVID-19 pneumonia into high probability, intermediate and low probability. So, even though we lost some opportunities during the COVID period, we used that time productively to build more products. We now have five products that we need to focus and sell.
Terence: There’s no distinction between work and home anymore. Honestly, we’ve gotten so much done from a work perspective and there’s still so much more to do. I don’t think we’ve worked harder, between the three of us, than we have over the last 16 months.
Even though we so often work hard and crazy hours, we always take time at lunchtime, where we just share a meal together. And I think that is good.
Do you know what you will use the Cisco Global Problem Solver $50,000 USD in prize money for specifically?
Jaishree: One of the things it will allow us to do is scale. The prize money will also allow us to access more markets and to get regulatory approval. We already have the South African regulatory approval, through SAHPRA, and we now want to get the FDA and CE marking on our products which will allow us to market our solution in the US, the UK, and elsewhere.
How will winning the Global Problem Solver’s Second Runner-up prize help you advance your business?
Jaishree: I think it already has. Just posting it out there has attracted a lot of attention. We’ve just started our capital raise and the coverage brought us to the attention of two potential investors, so the timing couldn’t have been any more perfect.
I also think that there are wonderful opportunities that may come down the road from working with Cisco. Possible opportunities to get together with some of the other winners that I was amazed and blown away by. That’s one of the upshots of the challenge that you guys may not realize you’ve made possible; the potential collaboration between some of the finalists and winners, working together for a greater good.
What advice do you have for other social entrepreneurs?
Andrei: Don’t give up. If you have an idea, just get it done, get it out.
Terence: When you’re driven by a strong passion and a strong purpose, and it’s a good purpose that has positive impact, that’s what’s going to drive you to do incredible work. You’re going to do groundbreaking work because you’re so driven by that inner passion and purpose for what you’re building, what you’re trying to achieve. You’re going to have bends and other cars in the road, but those don’t become such big distractions anymore if you have a strong purpose and vision.
Jaishree: I agree with Terry and Andrei. That’s exactly what gets me through it. Also, on the product side and building the solution, we’ve always continued to build, measure, and learn. We are very cognizant of making sure that this product works well. We measured it, with constant iteration. That is important, to always find your product, measure it, learn from it, and then make it better.
It’s important to build the right culture in the team and I’m so glad we had the opportunity to do that before COVID as it is difficult to build remotely. We had a good strong foundation and good culture in our company that we can and will continue to grow from.
More from Cisco Systems, Inc.