Osteoporotic Precise Screening Using Chest Radiography and Artificial Neural Network: The OPSCAN Randomized
Chin Lin1, Dung-Jang Tsai1, Chih-Chia Wang1
1From the Medical Technology Education Center, School of Medicine, National Defense Medical Center, Taipei, Taiwan, ROC (C.L.); Department of Artificial Intelligence (C.L., D.J.T., W.H.F.), Department of Family and Community Medicine (C.C.W., Y.P.C., J.W.H., W.H.F.), and Division of Cardiology, Department of Internal Medicine (C.S.L.), Tri-Service General Hospital, National Defense Medical Center, No. 325, Sec. 2, Chenggong Rd, Neihu District, Taipei TW 114, ROC; School of Public Health, National Defense Medical Center, Taipei, Taiwan, ROC (C.L., D.J.T.); and Department of Statistics and Information Science, Fu Jen Catholic University, Taipei, Taiwan, ROC (D.J.T.).
Artificial intelligence (AI) identified high-risk individuals for osteoporosis screening. AI-enabled chest radiograph screening significantly increased osteoporosis detection rates in this population compared to usual care.
Area of Science:
- Radiology
- Artificial Intelligence
- Osteoporosis Screening
Background:
- Osteoporosis diagnosis is challenging due to asymptomatic presentation.
- Screening high-risk populations is crucial for early detection.
Purpose of the Study:
- To evaluate the effectiveness of dual-energy x-ray absorptiometry (DXA) screening for osteoporosis.
- To assess an artificial intelligence (AI) model for identifying high-risk individuals using chest radiographs.
Main Methods:
- A randomized controlled trial included participants aged 40+ who underwent chest radiography.
- High-risk participants identified by AI were randomized to a screening group (offered DXA) or a control group (usual care).
- Logistic regression analyzed the difference in new-onset osteoporosis between groups.
Main Results:
- AI identified 12.1% of 40,658 participants as high-risk.
- The screening group showed significantly higher osteoporosis detection rates (11.1% vs 1.1%; OR, 11.2; P < .001).
- AI-identified high-risk participants not meeting formal DXA criteria had substantially increased odds of osteoporosis diagnosis (OR, 23.2).
Conclusions:
- AI-enabled chest radiograph screening effectively identifies high-risk individuals for osteoporosis.
- This approach significantly improves osteoporosis diagnosis rates in targeted populations.
- AI facilitates earlier and more effective osteoporosis detection, especially in those not meeting traditional screening criteria.
More Related Videos
06:59Author Spotlight: An Economic and Efficient Method for Quantitative Evaluation of Bone Microarchitecture in a Murine Osteoporosis Model
Published on: September 8, 2023
09:02Cortical Bone Assessment Using Ultrasonic Guided Waves: A Reproducibility Study in a Healthy Population
Published on: January 31, 2025
