M42 Reveals Landmark Findings In AI-Powered Tuberculosis Screening Study

M42, a leader in health technology, AI, and genomics, has revealed significant findings from a large-scale study on AI-driven tuberculosis (TB) screening. This research, conducted with the Capital Health Screening Centre (CHSC) in Abu Dhabi, is published in npj Digital Medicine - Nature. It stands as one of the largest real-world validations of an AI healthcare solution, analysing over one million chest X-rays (CXRs) to assess AI's effectiveness and scalability in TB screening.

The study evaluated M42's AIRIS-TB model, designed to streamline routine TB screenings. This allows radiologists to focus on complex or urgent cases. AIRIS-TB achieved an Area Under the Receiver Operating Characteristic Curve (AUROC) of 98.5%, indicating its high efficiency in triaging CXRs for TB. This model reported a 0% false-negative rate for TB-specific cases across the dataset, highlighting its safety and reliability.

According to the World Health Organisation (WHO), approximately 10.8 million people contracted TB and 1.25 million died from it in 2023. Recent reports indicate TB has returned as the leading cause of death from a single infectious agent globally. M42’s AIRIS-TB demonstrates AI's potential for real-world impact by safely automating up to 80% of routine CXR assessments. This reduces radiologist workload, minimises human error risk, and offers cost efficiencies in high-throughput low-prevalence settings.

Currently, reviewing CXRs is labour-intensive and prone to errors, potentially causing missed or delayed diagnoses. A previous study noted a 26.6% increase in missed findings when radiologists doubled their annotation speed and more errors after nine hours into their shift.

AIRIS-TB consistently performed well across various demographic groups, including differences in gender, age, HIV status, income levels, and populations spanning six WHO regions. This highlights the model’s robustness and fairness across diverse global populations. These results underscore AIRIS-TB’s potential to enhance clinical workflows significantly and drive earlier TB screening worldwide.

Dimitris Moulavasilis, Group CEO of M42, stated: "This landmark study marks a pivotal moment in the potential power of AI in the global fight against tuberculosis. Our AIRIS-TB model stands as a compelling testament to the unmatched accuracy, safety and scalability that AI can deliver."

Advancing Healthcare Delivery

Dr. Laila Abdel Wareth, CEO of Capital Health Screening Centre, commented: "The outcomes of this study reaffirm that AI models like AIRIS-TB can not only match – but safely surpass – human-level accuracy and efficiency in clinical practice." By automating routine screenings with precision, radiologists can focus on complex cases, enhancing diagnostic capacity.

The study underwent rigorous peer review by the Department of Health – Abu Dhabi to ensure transparency and accountability. Its publication reinforces M42's position as a pioneer in AI-led health solutions while solidifying the UAE’s role as a hub for medical innovation.

The full study is available through npj Digital Medicine.

With inputs from WAM