Read the Conversation
EF: Could you elaborate on your journey from radiographer to digital healthcare innovation entrepreneur?
KM: As a radiographer and mammographer, I specialize in breast cancer imaging and lecture postgraduates in imaging technology at the University of Pretoria. During my work in the field, I noticed specific cancer types, especially lobular carcinoma, being frequently overseen during imaging diagnostics. Its pathology fascinated me, and I wondered why we kept missing it or picking it up late.
Regarding the patient journey, I noticed a massive gap between our work in diagnostics and patient care during treatment. To improve the patient journey, I started a nonprofit seven years ago. Initially, we held tea events for about 40 ladies diagnosed with breast cancer in the Pretoria area. Now, we are engaged with over 200,000 women and educating them on the disease nationwide. The initiative has expanded, offering resources in all 11 official languages in South Africa, showing our commitment to raising awareness across society.
After a decade of working in practice, I quit my job and started researching breast cancer solutions and diagnosis from a different angle. I discovered MATLAB and taught myself programming languages, in which I wrote a basic algorithm to discern between different types of breast cancer. That was my lightbulb moment, as I knew I had discovered something massive. Thanks to the help of a patient who studied machine learning and artificial neural networks, we could improve the algorithm that set the foundation of MEDSOL. We worked together to apply for funding and product development. He is now one of the directors of the company.
Our first product was launched at the Unjani Clinics. Our AI system allows fast-track diagnostics in rural settings and reduces the patient load. Thanks to our smart system and our task teams present at the clinics; attention can be put on the patients who need surgical intervention. Hospitals can be more assertive when it comes to treatment, and the burden on the public healthcare system can potentially be significantly reduced.
However, it was not enough to just throw out a new tool to clinics without an education for the community. Our partnership with Astra Zeneca made us realize the importance of our nonprofit initiative when it comes to bringing people to the hospitals in the first place.
We are working together on building patient referral blueprints to guide them from their house to the point where they are diagnosed. We have even ensured care packs for patients and work constantly with the community to make sure patients are no longer lost in the system during their treatment.
EF: Healthcare professionals are very sceptical when it comes to AI. How are you engaging with healthcare professionals to promote and educate them about your project?
KM: There is a misconception that AI in radiology is a new thing. This technology has been around for over 30 years in the form of a basic computer-aided detection and segmentation algorithm. It has since been expanded. During our engagements with healthcare professionals, we have made it clear that the purpose of our product is not to replace anyone in the system. We are helping to fast-track and improve turn-around time. We are allowing general practitioners, through formal training initiatives, to advance their practice to develop point of care ultrasound skills.
Radiologists make use of AI in all forms of imaging. It is called second reader output. A second opinion is crucial in the medical profession and gives them reassurance.
Our AI system was exclusively trained on carefully curated data repositories, comprising exactly 40,000 histologically confirmed images sourced solely from the South African practice accredited by the American College of Radiology. This equips our users with a virtual specialist boasting over 25 years of invaluable experience, ready to assist them in cases where uncertainty arises.
EF: How does Medsol drive innovation in South Africa?
KM: We are starting at a small scale. Currently, we are only in the Gauteng region, but the plan for the A-Catalyst network is to expand to other provinces and, ultimately, to Kenya and Egypt. One of my biggest dreams when we initially started was to have mobile clinics with our AI system. The thought of having mobile clinics across Africa and being able to reach more people would be massive. This wish came true when the AstraZeneca branch in Egypt contacted us and told us that the local authorities wanted to start mobile clinics that would use our innovation.
Our AI is not limited to breast cancer diagnoses. We have started initial R&D for wound care detection, Wound AITM to be able to assess the progress during wound care at rural sites.
We are also working on chest imaging, specifically for the mining industry, Chest AITM. Very few radiologists specialize in interstitial lung diseases for the mining industry. While mines often have on-site X-ray facilities, they typically lack dedicated specialists to interpret the results. Our mission is to bridge this gap by enhancing the accessibility of point-of-care diagnostics. Our primary focus is identifying high-risk X-ray findings, which will be referred for in-depth examination. This approach aims to alleviate the strain on the workforce and public hospitals.
What sets our system apart is its adaptability. Our algorithm is highly versatile and capable of being trained on a wide range of categories. We can expand and refine its capabilities over time, tailoring it to new domains and applications. We are enthusiastic about the potential for further expansion into additional areas of impact.
EF: Are you planning further collaboration with other companies, and how do you choose the partners you would like to collaborate with?
KM: As a start-up, we are always looking for more companies to engage with. We want partners who focus on improving screening and are passionate about trying new things. This is why we engaged with Unjani Clinics and AstraZeneca. We are also in early talks with Roche because they want to promote and improve screening procedures. We want to engage with like-minded people whose passion and need for social impact align with ours.
EF: What advice would you give to entrepreneurs looking to make a mark in the fields of technology and healthcare?
KM: Find your passion and pursue it. Be prepared and take every opportunity to learn.
Three years ago, I quit my job and left my comfort zone. A saying states that if you choose a job you love, you will never have to work a day in your life. I found my passion three years ago and have not worked since then.
EF: Where do you see yourself and your company 10 years from now?
KM: I envision a future where we have hundreds of employees, with a particular emphasis on empowering more young women to excel in coding. I want to witness the widespread implementation of our solution in all 120 Unjani Clinics, extending our reach across the entire African continent.
A decade from now, I want to review breast cancer statistics and observe a drastic reduction, acknowledging that our contributions played a crucial role in driving this remarkable transformation.
EF: Do you have a final message for our readers?
KM: Funding is always a critical aspect. I want to encourage other startups, especially in South Africa, to keep trying and applying for opportunities.
