Research epidemiologist and CEO of THINK, Dr Kristina Wallengren.

• South Africa is among the top five countries with the highest burden of TB. 
• However, the government recently announced it would be slashing its budget for TB research grants.
• Now, a researcher plans to use AI to analyse TB data.

Machine learning and artificial intelligence (AI) may become a key weapon in fighting South Africa's tuberculosis scourge – which sees as many as 300 000 new infections and more than 50 000 deaths per year.

World Health Organisation data places South Africa among the top five countries in the world with the highest burden of TB, with more than 500 cases per 100 000 population.

Despite this, Health Minister Joe Phaahla announced last week that there would be a reduction of R1 billion in HIV and TB research grants.

He said the government was under financial pressure, and the department would contribute to reducing government spending.

Dr Kristina Wallengren, research epidemiologist and the CEO of THINK, a non-profit organisation committed to improving the lives of those affected by TB and HIV globally, said there was limited access to healthcare for those affected by TB.

She said South Africa had one of the highest HIV prevalence rates in the world, adding:

Wallengren added that they had received a grant from the Bill & Melinda Gates Foundation, which they would use for AI data analysis and interpretation to support evidence-based decision-making.

"The Department of Health does not have enough people to visit clinics and hospitals and then write a report and recommendations on managing TB. With AI, our work will be used by programme managers and coordinators to improve the TB programme," she said.

She said AI would help to assess the effectiveness of interventions and prioritise resources in a timely fashion, based on locally available data.

"THINK's AI hub will contribute to building AI literacy and data analysis and interpretation in the healthcare sector. With AI-facilitated decision support, the Department of Health can close the gap in the healthcare cascade and end TB by effectively utilising resources and interventions across all levels of the public health system, she said.

READ | 'We can make vaccines that protect for a lifetime' – Bill Gates

Meanwhile, Dr Fundile Nyati, CEO of Proactive Health Solutions, said TB was one of the reportable diseases in South Africa. 

"This is an excellent development, and it will help well-developed countries like South Africa regarding surveillance epidemiology systems. Using ChatGPT in TB will help to fast-track decision-making at a clinical level. It will augment epidemiologists' skills to get answers quickly," he said. 

Nyati said it was still crucial for tools like ChatGPT to be overseen by qualified professionals.

Dr Bandile Hadebe, a distinguished leader at the intersection of healthcare, technology and policy, said the application of technology in the public health sector made complete sense in managing big data.

"This artificial intelligence perfectly applies to TB from the track and trace perspective. There is a massive possibility of use of technology in fighting the TB scourge," he said. 

He said it was concerning that the data available on ChatGPT had a time stamp.

Hadebe added: 

She said people fell away at every layer of the healthcare cascade, from screening and diagnosis to treatment and care.

"It's like peeling an onion where the outer layer represents everyone infected by TB. Of those, not all recognise the symptoms in time or seek care; thus, a layer of the onion is removed. Not everyone is diagnosed for those seeking care, and those diagnosed are started on treatment," Wallengren said.

Wallengren, who said she had been researching TB in the country, added that Covid-19 had reversed the decades of hard work and progress towards fighting TB.

"Many countries reallocated healthcare resources, including funds, medical staff and facilities, to address the Covid-19 pandemic. This meant fewer resources were available for TB diagnosis, treatment and prevention. Many TB clinics were temporarily closed, and routine check-ups, diagnosis and treatment were interrupted," she said.

She said poverty, overcrowding, informal settlements and malnutrition could increase the risk of TB transmission as well as the progression of latent TB infection to active TB.