Welcomed by nearly anybody who believes wellness and prevention are the best way to build healthy populations (though perhaps less healthy private practice bank balances), AI and precision medicine are two vehicles that help us do just that. With a database that makes Discovery Health green with envy, the NHI’s Dr Nicholas Crisp would be one sure-fire AI advocate, but it will first take some building across the current patient care, public/private divide. Perhaps its time to break out of our self-protected curative treatment silos and engage with the power of virtual reality…Read on for input from some top experts:
Artificial Intelligence’s burgeoning success had added a new term to the medical lexicon – one that sums up its’ almost incredible utility in the Fourth Industrial Revolution – ‘Previvor”: the person who AI-assisted technology has pre-diagnosed as being in almost certain danger of developing a serious or life-threatening condition. No doctor on their own can do so with anywhere near equal accuracy.
Most doctors will concede that they know only fragments of medicine well – well, GPT4, the latest iteration of ChatGPT, knows a lot of medicine well. It’s an even more powerful language model than ChatGPT because it’s multi-modal, works with images (including video), and speech, and allows its users to look at unstructured text.
Too few of us use either iteration. It you take the entire internet, Wikipedia, a few hundred billion books and feed all that data into AI computation, you’ve got GPT 4.
Unsurprisingly, in healthcare, AI programs now conduct clinical diagnosis with a better accuracy rate than their human “colleagues.”
Associate Professor Deshen Moodley, the first occupant of UCT’s new SA Research Chairs Initiative in AI Systems, says that when it comes to some doctors’ resistance to both precision medicine and AI, ‘the train has already left the station”.
“Physicians will have to scramble to catch up. If they don’t adjust, they’ll be left behind. People have access to a whole lot more knowledge today and patients will be the future drivers of change. A lawyer friend of mine told me in surprised tones the other day that his Apple watch said something was wrong with his heart. He went to see his cardiologist who confirmed it and said he was lucky to have come in,” he says, to illustrate.
With smart phones, computers and digital technology giving us the ability to maintain our own health and well-being there’s a fast-moving shift of power away from doctors, plus the potential to slash healthcare budgets by half within 20 years, Moodley says.
“I’m no health economist, but those are the figures. The measures should be health and well-being and not patient numbers in hospitals or operations performed,” he adds.
Besides open-source patient access to healthcare knowledge, the tools are already here to assist doctors with better predictive diagnoses and preventative treatments.
Here’s how
Take AI-powered
coronary artery scan technology, developed by an American company called Prenuvo. Among many iterations that will potentially revolutionise healthcare through its preventive approach to diseases, the AI-powered scan uses magnetic resonance imaging (MRI) to flag and escalate potential health problems – ‘seeing’ potential complications the radiologist and/or cardiologist simply cannot.
So, before the symptoms of any disease emerge, the technology predicts it, enabling physicians to figure out prevention methods. It analyses every part of the body and the organs from head to toe to detect any anomalies from cancer to liver stones and kidney diseases – all forms of pathologies. There are no invasive procedures or radiation, and the results are made available to the patient within 48 hours of the examination.
Then there’s the AI-powered
coronary artery scan technology developed by the Fountain
Life Health Technology Company in the United States. It can detect heart pathologies, especially heart attacks, a decade before the symptoms surface. That transforms healthcare from a reactive to an initiative-taking domain. The AI coronary artery scan (CAS) is similar to the coronary computed tomography angiography (CCTA), in use now for several decades to diagnose heart diseases. However, the difference is that the CAS employs not only the result analyses of radiologists and cardiologists but adds in AI to scrutinise the results of the scan.
AI analyses ensure that plagues previously concealed from the human eyes are seen.
SA research links
Dr David Jankelow, a cardiologist working out of the Linksfield Park Medical Centre in Johannesburg and a past president of the South African Heart Association, is working on AI with the Mayo Clinic in the United States. He confirms that the Mayo Clinic has developed an algorithm that works with an ECG machine to predict future cardiac disease – before it clinically manifests. Together they’re seeing amazing results predicting atrial fibrillation, the biggest cause of stroke, aortic valve disease, cardiomyopathy, and other heart dysfunctions.
Jankelow explains: “They took a cohort of some 53 000 patients and fed their data into an algorithm paired with an ECG and ultrasound. Then they took a different data set of patients and fed their ECG readings in and asked who has a weak heart pump. The accuracy was 93%. They followed all the computer false positives and the truly AI ECG tested patients for several years. The former had a four-fold chance of developing a weak heart and the latter remained well.”
He said an entirely new term had been added to medicine’s lexicon: ‘previvor.”
Jankelow is excited about screening for valvular, mitral, and rheumatic heart disease, all of which are abundant in South Africa, providing rich data to feed into the Mayo Clinic AI programme.
In 2013, an estimated one million deaths were attributable to cardiovascular disease (CVD), in Sub-Saharan Africa (SSA), constituting 5.5% of all CVD-related deaths globally and 11.3% of all deaths in Africa. Between 1990 and 2013, SSA was the only geographical region of the world in which the number of CVD-related deaths increased.
Prof Athol Kent, a UCT-based medical education innovator and obstetrician/gynaecologist, writing in his popular Journal Article Summary Service, (JASS), observes; “AI will move us forwards and Luddite responses will merely discredit us. We, the doctors, will have to figure out how AI works to our advantage and stay informed as to how this is best achieved. We need more healthcare professionals, we need different types of healthcare professionals and maybe these will be a mix of humans with communication skills and AI technology, we need better ways reaching and responding to our patients, we need more time with those we care for, and we need better systems to run our practices and hospitals. All these possibilities can be facilitated by AI to the clinical benefit of us all.”
Jankelow adds: “It’s not prime time yet. There’s a large gap, some call it the digital divide. In other words, from research and development of this technology to implementation in clinical practice. But it’s hotting up, there’s so much going on behind the scenes. It will streamline workflow in a hospital – we’re not doing it yet, but that may well come first. AI will also standardise procedures and predict who will suffer an adverse event. The dream for me is that it could help democratise healthcare. There are areas in SA where there are no doctors, nursing sisters or ECG machines. As far back as 2016 already we had one cardiologist per 260 000 people. Brazil had one in 23 000, ten times more than us. That hasn’t changed much,” he added.
How AI evolved
Tracing the recent evolution of AI, Dr Isaac Kohane, of Harvard University, a pioneer in the AI field, and a paediatric endocrinologist who leads Harvard’s Bioinformatics Department, says things began changing in 2012.
“We were already pretty impressed with it, but around 2018, we started seeing, in the medical literature, the consequences of these convolutional neural networks — that they could actually detect changes in images that were perhaps imperceptible to humans, which would allow them not only to diagnose retinopathy in the back of the eye (for example), but could tell you their BP, the sex of the patients, their age, and what other diseases they had.
Jankelow cites several studies predicting that by 2040 there will be 640 million diabetics world-wide, with a third of them women suffering some form of diabetic eye disease. The global shortage of ophthalmologists, narrowed down to South Africa, comes in at six ophthalmologists per one million people.
“Some form of computer analysis will hugely ease the burden on the healthcare system – and that’s just one discipline,” he says.
Elaborating on the evolution of IT, Kohane adds: “So, there were a bunch of these mostly image-based applications in 2018, but also some applications related to time series and health records, which had impressive performance. What was different, what was characteristic of them, was that they were purpose-built. You train them for a specific purpose: diagnosing retinopathy, predicting time to readmission.
“These were programs that not only were purpose-built but, because of that, you could easily evaluate them and assess their accuracy for a specific task.”
GPT stands for Generative Pretrained Transformer, a natural language processing tool that allows you to have a personalised conversation with an AI bot. It’s the fastest uptake of any technology in the history of humankind, according to Jankelow, with one hundred million users globally already.
“It passed the USA medical licensing exams with no cramming – unlike human students. We need to remember, it’s not a replacement for specialised knowledge, critical thinking, and ethical consideration in the practice of medicine. But it’s the next step in the Fourth Industrial Revolution where all processing in term of search tools is going. Every search engine will have this.”
Quoting Dr Eugene Stead, who paved the way for cardiac catheterization in the 1940’s and led the concept of a computerised textbook for medicine, he said AI was, “the art of making machines smart,’ and predicted it would shortly render radiologists redundant.
“If you’re a radiologist, you’re already over the edge of the cliff – you just haven’t looked down yet. There’s no ground underneath. In five years, deep learning is going to do the job better than radiologists. They should stop training them now,” Jankelow quoted.
He disagreed, however, with those who said AI was ‘some kind of magic bullet’: “The answer is no in the short term, but likely in the medium to long term. Imagine a complete overview of patient data and the reduction in medical errors? Solving medical problems and recommending individualised treatment will become the norm.”
AI – equity and pandemics
Asked whether South Africa could move to equity with AI, given that it carries the world’s largest Gini-co-efficient, Prof Moodley replied: “There are two big issues. One is access to AI (beyond just health) – that drives our AI group at UCT. We need open-source platforms, otherwise the cost goes up. We need to mobilise the use of AI for open-source applications and reference implementations and platforms to reduce cost. The other is that if you look at our current systems, we’re only using about 10 percent of them in terms of innovative AI possibilities on the preventative side. Google Search, Chat GPT and GPT 4 have knowledge that can empower individuals to start managing their health. The only difference between myself and entrepreneurs is that our public health platforms will be freely available, whereas they want to make money off it.”
He said that with Covid, for example, AI had the ability to play out all the response models in pseudo reality.
“It will not just follow an outbreak in real time but see it spatially and give you a completely different real-time decision-making tool. You can rewind and see what happened in the past, scrutinise, and say this led to that, whether it be Covid or any communicable disease. Now, with that simulation and modelling you’ve got a much clearer idea of which intervention will do what. It’s radical, super-charged for any public health intervention wherein you can see the possibilities play out.”