Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 31 min ago
Peer-reviewedCC BYHealthcare (Basel, Switzerland)

Real-World Validation of an AI-Based Opportunistic Osteoporosis Risk Stratification Tool Using Chest Radiographs in a Male-Predominant Cohort.

  • Kao-Chang Tu1,2
  • Sheng-Chou Wu1
  • Han-Ting Shih1,3
  • Cheng-Hung Lee1
  • Kun-Hui Chen1,4,5
  • Chia-Hsien Hsu2,6,7
7 affiliations
  1. 1Department of Orthopaedics, Taichung Veterans General Hospital, Taichung 40705, Taiwan.
  2. 2Doctoral Program in Tissue Engineering and Regenerative Medicine, National Chung Hsing University, Taichung 402202, Taiwan.
  3. 3Department of Industrial Engineering and Enterprise Information, Tunghai University, Taichung 407224, Taiwan.
  4. 4Department of Post-Baccalaureate Medicine, College of Medicine, National Chung Hsing University, Taichung 402202, Taiwan.
  5. 5Department of Computer Science and Information Engineering, Providence University, Taichung 43301, Taiwan.
  6. 6Institute of Biomedical Engineering and Nanomedicine, National Health Research Institutes, Zhunan, Miaoli County 35053, Taiwan.
  7. 7Institute of NanoEngineering and MicroSystems, National Tsing Hua University, Hsinchu 30013, Taiwan.
Read full textPDFDOICite

Abstract

Background: Osteoporosis is a major public health concern and a leading cause of fragility fractures, yet screening with dual-energy X-ray absorptiometry (DXA) remains limited by accessibility and utilization. Opportunistic artificial intelligence (AI)-based analysis of routinely acquired chest radiographs (CXRs) may facilitate osteoporosis risk stratification. This study aimed to independently validate a regulatory-approved deep learning model (VeriOsteo™ OP) in a real-world, male-predominant cohort. Methods: We retrospectively analyzed 2336 patients who underwent both CXR and DXA at a medical center in 2023, including 1688 males (72.3%) and 648 females (27.7%). Osteoporosis was defined as a DXA T-score ≤ -2.5. AI performance was evaluated using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and area under the receiver operating characteristic curve (AUC), with sex-specific subgroup analyses. Results: The AI model achieved an accuracy of 76.97%, sensitivity of 69.67%, specificity of 84.64%, PPV of 82.66%, and NPV of 72.65%. The overall AUC was 0.854 (95% CI, 0.839-0.868). The AUC was 0.825 (95% CI, 0.794-0.854) in males and 0.849 (95% CI, 0.831-0.866) in females. PPV was lower in males than in females (60.00% vs. 83.39%), while other performance measures were broadly comparable. The optimal AI cutoff also differed between males (≤-1.9) and females (≤-2.5). Conclusions: The AI model demonstrated good discrimination for opportunistic osteoporosis risk stratification using CXRs in a real-world cohort. Its moderate sensitivity and lower PPV among males highlight potential sex-related differences in AI-assisted osteoporosis assessment. The model may be most appropriately used as an adjunctive risk-stratification and DXA-prioritization tool to identify individuals who may benefit from subsequent DXA evaluation rather than a standalone diagnostic test.

Related

  1. Federated learning for privacy-preserving opportunistic screening of lumbar-spine DXA-defined osteoporotic-level bone mineral density using chest radiographs: A multi-center independent validation study
    Bo-Ru Lin, Ming-Yueh Huang, Wei-Chung Sun +4
    Osteoporosis and sarcopeniaPeer-reviewedCC BY-NC-NDFull textGeneral medicineEurope PMC
  2. Cost-utility analysis of deep learning-based opportunistic osteoporosis screening using chest radiographs: implications of diagnostic thresholds and confirmatory testing.
    Bo-Xiang Shi, Dung-Jang Tsai, Chin Lin +3
    Archives of osteoporosisPeer-reviewedCC BYFull textGeneral medicineEurope PMC
  3. AI-Driven opportunistic osteoporosis screening: the skeletal failure paradigm for trustworthy vertebral fracture triage
    Mohamed Ragab Nouh
    The Egyptian Journal of Radiology and Nuclear MedicinePeer-reviewedCC BYFull textMedical imaging & radiologyDOAJ
  4. Opportunistic computed tomography to identify osteoporosis in primary hyperparathyroidism using sestamibi SPECT/CT scans
    Gary Ka Wai Chan, Xi Xiong, Yan Ho Hui +2
    Journal of the Endocrine SocietyPeer-reviewedCC BYFull textSurgery & anaesthesiaEurope PMCPMC
  5. Opportunistic computed tomography-based vertebral attenuation for osteoporosis screening
    Irfan Atik, Aibek Tashimbekov, Enes Gul +2
    Revista da Associacao Medica Brasileira (1992)Peer-reviewedCC BYFull textSurgery & anaesthesiaEurope PMCPMCEurope PMC

Cite

BibTeXRIS
BibTeX
@article{tu2026realworld,
  title = {{Real-World Validation of an AI-Based Opportunistic Osteoporosis Risk Stratification Tool Using Chest Radiographs in a Male-Predominant Cohort.}},
  author = {Kao-Chang Tu and Sheng-Chou Wu and Han-Ting Shih and Cheng-Hung Lee and Kun-Hui Chen and Chia-Hsien Hsu},
  journal = {Healthcare (Basel, Switzerland)},
  year = {2026},
  month = sep,
  doi = {10.3390/healthcare14193200},
  url = {https://europepmc.org/article/MED/42881638},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Real-World Validation of an AI-Based Opportunistic Osteoporosis Risk Stratification Tool Using Chest Radiographs in a Male-Predominant Cohort.
AU  - Kao-Chang Tu
AU  - Sheng-Chou Wu
AU  - Han-Ting Shih
AU  - Cheng-Hung Lee
AU  - Kun-Hui Chen
AU  - Chia-Hsien Hsu
PY  - 2026
DA  - 2026/09/28
JO  - Healthcare (Basel, Switzerland)
DO  - 10.3390/healthcare14193200
UR  - https://europepmc.org/article/MED/42881638
AB  - Background: Osteoporosis is a major public health concern and a leading cause of fragility fractures, yet screening with dual-energy X-ray absorptiometry (DXA) remains limited by accessibility and utilization. Opportunistic artificial intelligence (AI)-based analysis of routinely acquired chest radiographs (CXRs) may facilitate osteoporosis risk stratification. This study aimed to independently validate a regulatory-approved deep learning model (VeriOsteo™ OP) in a real-world, male-predominant cohort. Methods: We retrospectively analyzed 2336 patients who underwent both CXR and DXA at a medical center in 2023, including 1688 males (72.3%) and 648 females (27.7%). Osteoporosis was defined as a DXA T-score ≤ -2.5. AI performance was evaluated using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and area under the receiver operating characteristic curve (AUC), with sex-specific subgroup analyses. Results: The AI model achieved an accuracy of 76.97%, sensitivity of 69.67%, specificity of 84.64%, PPV of 82.66%, and NPV of 72.65%. The overall AUC was 0.854 (95% CI, 0.839-0.868). The AUC was 0.825 (95% CI, 0.794-0.854) in males and 0.849 (95% CI, 0.831-0.866) in females. PPV was lower in males than in females (60.00% vs. 83.39%), while other performance measures were broadly comparable. The optimal AI cutoff also differed between males (≤-1.9) and females (≤-2.5). Conclusions: The AI model demonstrated good discrimination for opportunistic osteoporosis risk stratification using CXRs in a real-world cohort. Its moderate sensitivity and lower PPV among males highlight potential sex-related differences in AI-assisted osteoporosis assessment. The model may be most appropriately used as an adjunctive risk-stratification and DXA-prioritization tool to identify individuals who may benefit from subsequent DXA evaluation rather than a standalone diagnostic test.
SN  - 2227-9032
ER  -