AI Innovations: What This South African Doctor Is Making Possible

Published 11 days ago
Tiana Cline
Medical technology, doctor use AI robots for diagnosis, care, and increasing accuracy patient treatment in future. Medical research and development innovation technology to improve patient health.
Medical technology, doctor use AI robots for diagnosis, care, and increasing accuracy patient treatment in future. Medical research and development innovation technology to improve patient health. (Getty Images)

The fact that Africa has a shortage of clinical oncologists has been widely publicized. One international medical journal estimated that Nigeria, for example, has around 70 oncologists to serve a population of 213 million people. In South Africa – and in Johannesburg, in particular – it’s been reported that Gauteng’s health department has allocated millions to cancer care to assist the main hospitals that have become overwhelmed with patients needing critical treatment. And while there is definitely a skills shortage, Dr Duvern Ramiah, the head of radiation oncology at the University of the Witwatersrand and Charlotte Maxeke Johannesburg Academic Hospital explains that no one had really done a deep dive into what is actually holding up access to these life-saving interventions.

“The main bottleneck with radiation and oncology treatment is a lack of radiotherapy planning. That’s the irst step to patients get- ting radiation,” he says. With prostate cancer patients waiting around 5 years for radiotherapy and breast cancer patients waiting around 18 months, Ramiah came to the realization that there wasn’t just a shortage of equipment; the unit wasn’t using the equipment that it had efficiently. “The unit wasn’t creating enough radiotherapy plans for these patients to be treated in time,” he adds. When a cancer patient requires radiation, radiation is prescribed by a radiation oncologist. What happens next is that the patient is scanned and goes through the long process of radiotherapy planning. “If a patient has prostate cancer, the doctor will say (for example), this person needs external beam radiation. I’ll see him and make that call. Ater that, he’ll go have a CT scan which is done in the position that he’s going to have radiation in,” explains Ramiah.

Once Ramiah receives the CT scan, he’ll have to manually draw in all the areas that need to get radiated, like the prostate and lymph glands, as well as key areas to avoid, like the bladder. “It’s a long process, and only once that’s done, I’ll send the scan off to a dosimetrist or planning radiotherapist,” he continues. Once a plan is created, it still needs to be approved by the radiation oncologist to ensure it meets the prescription requirements.

Advertisement

Depending on the type, stage and biological behavior, some cancers metastasize quicker which means these plans are needed pretty quickly so that they don’t become outdated. Ramiah turned to AI to create a new process where part of this journey could be automated. “So, instead of a doctor drawing in all of those organs on scans, we’ve got a convolutional neural network (CNN) running that will contour target areas. The time a doctor is spending doing this process is very mundane when it can just be automated. It reduces the time gap massively,” he says. Another way Ramiah has reduced the bottleneck is by using remote radiotherapists in America to create a detailed radiotherapy plan that oncologists at Charlotte Maxeke can implement. What Ramiah also found is that with the planning process, there is a lot of rep- etition. A plan for breast cancer can take up to two hours but around 70% of what is needed is actually the same from one patient to the next. “So, we’re also developing a script to automate some of that,” he adds.

With the help of AI, Ramiah has cut Charlette Maxeke’s two-month waiting time for radiotherapy plans down to between 48 and 72 hours. One of the most important parts of this AI journey is that the technology system Ramiah is piloting at Charlotte Maxeke is vendor-agnostic. So, because it isn’t tied to a specific brand or model of radiotherapy machine, it could potentially be implemented in any healthcare facility anywhere in the world.

Loading...

“You can go to any unit in Africa with this solution, regardless of their technology infrastructure,” he says. “Unfortunately, within cancer, we get stuck in a lot of innovative things like gene therapy, which is a million-rand treatment but not accessible to everybody. What’s interesting about AI tech is that it increases access. It’s an enabler, which is why I push so hard for it in my department.”

Advertisement

Loading...