Osteoporosis is a systemic disease characterized by bone loss, particularly affecting the elderly and postmenopausal women. It leads to increased bone fragility and a higher risk of fractures. Although Dual-Energy X-ray Absorptiometry (DXA) is currently considered the gold standard for measuring bone mineral density in clinical practice, the high cost and limited availability of DXA equipment result in many potential patients missing opportunities for early detection and preventive treatment.
To address this challenge, Dr. Kun-Hui Chen, Director of the Orthopedics Department at Taichung Veterans General Hospital, collaborated with Acer Medical to develop the “VeriOsteo® OP” AI-powered osteoporosis screening solution. Leveraging deep learning technology to analyze chest X-ray images, this system can quickly predict abnormal bone density within 30 seconds. VeriOsteo® OP has been certified by Taiwan’s Ministry of Health and Welfare as a medical software device and is now available to serve more patients.
AI-Powered Chest X-ray Analysis Enables Fast and Accurate Bone Density Detection
VeriOsteo® OP uses deep learning technology to analyze posteroanterior (PA) view chest X-rays. By identifying the 12th thoracic vertebra (T12) and the first lumbar vertebra (L1), it screens for abnormalities in bone density. At the core of this technology is an AI model capable of accurately interpreting skeletal structures within the image and inferring the patient’s bone density. It then converts the results into a T-score based on the World Health Organization (WHO) standards, with a T-score of -2.5 used as the threshold to determine the presence of abnormal bone density. Key advantages of the system include:
Automated AI Image Analysis: Trained on over 5,000 clinical cases, the deep learning model automatically detects and analyzes images of the T12 and L1 vertebrae to infer bone density values.
Rapid Diagnosis and Reporting: Analysis and results for a single chest X-ray image are generated within 30 seconds, providing immediate diagnostic information without requiring additional tests.
High Accuracy: Clinical trials show a sensitivity of 71%, specificity of 89.36%, and an AUC of 0.94, effectively supporting healthcare professionals in osteoporosis screening.
No Additional Equipment and Cost-Efficient Application: The system integrates directly with hospital PACS and HIS systems, enabling patients to receive bone density screening results during routine check-ups. It eliminates the need for high-end GPU processing, significantly reducing equipment costs for healthcare institutions.
Enhanced Screening Rates and Early Intervention Opportunities: Chest X-rays are commonly included in routine health exams, increasing screening coverage and facilitating early prevention and treatment.
Expanding Global Markets and Advancing AI Medical Device Applications
Dr. Kun-Hui Chen emphasized that with the accelerating global trend of population aging, the need for osteoporosis prevention and treatment continues to rise. The successful development and application of this technology aim to improve osteoporosis screening rates, enabling healthcare providers to identify high-risk groups earlier and reduce fracture-related medical expenses and social costs. Through close collaboration with Acer Medical, the technology has already expanded into Southeast Asian markets, with plans to further extend into Japan, the United States, Europe, and other international markets. This initiative seeks to promote the application and development of AI-powered medical devices globally.
Looking ahead, the collaborative team will continue to optimize the AI model to enhance its accuracy and broaden its scope of application. They plan to integrate the system with telemedicine and big data analytics to create a comprehensive bone health management platform.
Resource: 胸部X光識別密度異常 30秒檢驗骨鬆
