With Africa accounting for 23% of Tuberculosis cases, AI-powered Epi-control platform has improved the detecting process of more cases globally.
The Epi-control platform was developed by EPCON, which is a healthcare impact organisation that specialises in the use of AI to quantify health risks at high spatial resolution.
The Epi-control platform is providing local teams and healthcare workers with rapid insight into communities that have a high risk of TB transmission.
The World Health Organization (WHO) reported the fight against TB is showing slow progress because of limited resources to quicken the detection process.
However, the use of AI is showing promise in revolutionising the TB detection process.
The CEO of EPCON, Caroline Van Cauwelaert, pointed out that TB is challenging to detect in low-and middle-income countries (LMIC), especially with an over-stretched healthcare system.
She said AI use has transformed the ability to detect TB in African countries.
TB can remain asymptomatic, making its carriers often unaware of their condition.
Those with pulmonary TB can infect up to 15 other people through close contact over a year.
“Without proper treatment, two-thirds of those infected with TB will die. When TB can be quickly and effectively detected, it can be cured – antibiotics have been available since the 1950s,” said Van Cauwelaert.
A study published in the “Tropical Medicine and Infectious Disease” journal indicated that the AI capabilities of the Epi-control platform’s hot spot mapping model were determined to be 75% more accurate in predicting TB hotspots in Nigeria than traditional approaches.
The Epi-control platform is used to facilitate data visualisation, real-time monitoring, and predicting.
Using Bayesian machine learning, the Epi-control platform uses a hybrid genetic algorithm to process large-scale epidemiological and clinical data to predict disease spread and impact.
The AI model has supported projects across Pakistan, Nigeria, the Philippines, South Africa, the DRC and Guinea, which benefited more than two billion people.
“Using our Epi-control platform which integrates routine programme data with local socio-demographic and contextual information, we identified TB hotspots for targeted active case finding with impressive results.
“The TB positivity yields at these model-predicted hotspots were significantly higher than those at conventionally selected sites – 73% higher in Lagos, 95% in Ogu, 103% in Osun and 75% in Oyo,” said Van Cauwelaert.
EPCON partnered with Aquity Innovations to develop a TB risk predictive model for the Nelson Mandela Bay area, which increased detection rates from 0.2% in random screening to 0.9% using the model-based approach.
Aquity is a public health organisation that leverages evidence-based interventions to support the South African government’s efforts to develop cost-effective strategies aligned with the global End TB goals aimed at eliminating TB by 2035.
Director of health programmes at Aquity, Dr Sipho Nyathi, said: “With the Epi-control platform, we can now quantify the TB burden at the community level, allowing us to focus our efforts more precisely and ensure interventions are both targeted and cost-effective, which is critical in environments where resources are increasingly constrained.”
Van Cauwelaert said as an organisation, EPCON is committed to strengthening health systems, and supporting government and NGOs.
“This requires a holistic and integrated approach, involving all stakeholders including healthcare providers, patients, and public health officials.
“We are committed to expanding our AI’s capabilities to model a broader spectrum of diseases, including non-communicable conditions such as diabetes and cardiovascular diseases,” said Van Cauwelaert.
The Epi-control platform in the country drastically reduced the cost of finding undiagnosed TB cases from $1748 to $437, compared to conventional approaches.
The Star