Bangkok, Thailand – 8 May 2020: Internet Thailand Plc (INET) and IBM today announced the use of IBM Visual Insights AI vision technology and IBM Power Systems AC922 servers, the world-class high-performance computing system behind the world’s most powerful supercomputers, to help radiologists detect tuberculosis (TB) cases from chest X-ray images. The announcement marks the first use of computer vision technology in X-ray imaging interpretation in Thailand. The AI-powered CXR screening application, currently being used at 75 community and medical correctional hospitals across Thailand, can now help flag the TB symptoms, ensuring thousands of patients receiving timely detection, diagnosis and treatment.

Tuberculosis is an infectious disease and is a serious threat to human health - according to the World Health Organization (WHO), it is one of the top 10 causes of death, with around 10 million people fall ill with it each year. One-quarter of the world’s population is thought to be infected with latent tuberculosis and Thailand is one of the 14 worst countries hit by tuberculosis, with 108,000 occurring and reoccurring cases each year¹. But it’s also perfectly manageable and treatable - if more people are aware of and understand the disease, and with timely detection, diagnosis and treatment.

One of the biggest challenges in tracking the infection in Thailand is the reporting system. Though chest X-ray is one of the most common imaging exams used worldwide, Thailand is experiencing shortage of radiologists, which requires many smaller hospitals to send the X-ray images to the more equipped hospitals for X-ray images interpretation. Patients in the rural communities and the remote areas will have to wait for weeks to months before they can be diagnosed and received the right treatment. In addition, the inadequacy of skilled and experience radiologists has also affected the accuracy and completeness of identification of relevant findings and diagnosis of the patients.

“As Thailand’s most comprehensive cloud solution provider, INET is committed to ensure that the utilization of innovative technology will help to improve quality of life, productivity and efficiency for all businesses in the nation,” said Mrs. Morragot Kulatumyotin, INET’s Managing Director. “The combined strengths of INET and IBM in technological innovation and development, together with the domain expertise from radiologists, physicians and experts’ inputs from hospitals across Thailand truly represents the next frontier of a more accessible healthcare landscape in Thailand, made possible by disruptive technology.”

By training the AI system with over 40,000 images from hospitals and chest radiography models instructed by radiologists and physicians, the CXR screening app developed by INET has proven to be able to detect tuberculosis with 96% accuracy rate. Leveraging IBM Visual Insights, a visual recognition solution running on IBM Power Systems, allows INET to create visual models, then classify detect objects in chest X-ray images without requiring coding or deep learning expertise.

The security of IBM Systems as well as the ISO 27017 and ISO 27799 certification of the healthcare cloud and storage also ensure the patients of the appropriate personal data protection and application of security controls without compromising privacy or risk levels. The users will also receive efficient support from Thai experts.

“We are proud to collaborate with INET to bring this disruptive AI vision and high-performance computing technology, as well as secured world-class platform, to support radiologists and physicians and help tuberculosis patients to have access to quality diagnosis and timely treatment at the hospitals across Thailand,” said Patama Chantaruck, Vice President for Indochina Expansion and Managing Director of IBM Thailand. “This has truly reflected how Thailand’s healthcare landscape is evolving and leading the transformation at both national and international levels."

Radiologists can now simply upload the chest X-ray images to the CRX screening app, where the interpretation results will appear in seconds together with the confidence rating percentage. The technology allows doctors to place more focus on the potential TB cases to ensure timely and effective treatment care and measures. The use of AI-based CRX screening app is currently being extended to other 14 chest X-ray symptoms detection such as pneumonia, pulmonary edema and pneumothorax, and is expected to help thousands of patients in 300 hospitals across Thailand next year. The technology is also used today to train the medical students about the chest disease diagnosis and treatment.

############################

Reference:

[1] Anti-Tuberculosis Association of Thailand

Media contacts:

Paranee Reymondon
IBM Thailand Co., Ltd.
Tel: +662 273 4164 Email: paranee@th.ibm.com