AI Takes on Tuberculosis: New Diagnostic Tech Approved in South Korea (Image created by ChatGPT)
SEOUL, May 14 (Korea Bizwire) — Starting this June, South Korea will begin using artificial intelligence in clinical settings to assist with the interpretation of chest X-rays, aiming to enhance early detection of major pulmonary diseases such as lung cancer and tuberculosis.
The Ministry of Health and Welfare has designated the AI-assisted diagnostic tool as an “innovative medical technology,” authorizing its use for three years from June 1, 2025, to May 31, 2028. The tool, which analyzes chest radiographs for abnormalities, will be used on adult patients suspected of having lung conditions.
The AI system identifies five key abnormalities: nodules, consolidation, interstitial markings, pleural effusion, and pneumothorax. When a patient undergoes a standard chest X-ray, the AI algorithm analyzes the image and flags any abnormal findings, indicating their location to aid physician review.
While the Korean National Evidence-based Healthcare Collaborating Agency (NECA) has confirmed the technology’s safety, further data from real-world clinical environments will be gathered over the three-year period to evaluate effectiveness. The results will inform whether the AI tool becomes eligible for national health insurance coverage after a reassessment in mid-2028.
A Support Tool, Not a Standalone Diagnosis
Health officials emphasize that the AI system serves as a support tool for physicians, not a replacement. It aims to boost diagnostic accuracy by detecting subtle or obscured anomalies that human eyes might miss, and to help clinicians focus more effectively on critical cases, thereby improving workflow and efficiency.
Still, the final diagnosis remains with medical professionals, who must consider the patient’s symptoms, medical history, and other test results. The government directive explicitly warns against overreliance on the AI tool, noting the possibility of both false positives and missed findings.
Chest X-rays of patients with Wuhan pneumonia.
(Image courtesy of the New England Journal of Medicine )
Public Health Benefits — and Policy Challenges
Officials say the technology could contribute significantly to public health by enabling earlier diagnoses and reducing treatment costs. However, several challenges remain.
As the tool is categorized under “innovative medical technology,” it is not yet covered by national insurance, meaning patients may need to pay out-of-pocket. Additionally, effective AI learning requires large volumes of clinical data, raising concerns about patient privacy and data security, which the government says must be rigorously protected.
The AI service will initially be available only in certified medical institutions that submitted applications to NECA, and only physicians working at those institutions may use the tool during its provisional three-year evaluation period.
As the country continues to integrate AI into healthcare, the success of this pilot program will likely shape the future of medical AI adoption in South Korea’s national healthcare system.
Lina Jang (linajang@koreabizwire.com)
