Application of Artificial Intelligence for Osteoporosis Screening Using Chest Radiographs

Polasan  Santanapipatkul; Lakkana  Jirapong; Anchalee  Chumjam; Nantika  Wannasoupol; Rapeepat Narkbunnam

doi:10.33192/smj.v78i1.277295

Authors

Polasan Santanapipatkul Department of Orthopedic Surgery, Samut Sakhon Hospital, Samut Sakhon, Thailand
Lakkana Jirapong Department of Radiology, Samut Sakhon Hospital, Samut Sakhon, Thailand
Anchalee Chumjam Department of Otorhinolaryngology, Samut Sakhon Hospital, Samut Sakhon, Thailand
Nantika Wannasoupol Department of Radiology, Samut Sakhon Hospital, Samut Sakhon, Thailand
Rapeepat Narkbunnam Department of Orthopedic Surgery, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand

DOI:

https://doi.org/10.33192/smj.v78i1.277295

Keywords:

AI, BMD, osteoporosis, screening, chest radiograph

Abstract

Objective: To create the artificial intelligence screening osteoporosis tool (AISOT) to predict bone mineral density (BMD) using chest radiographs and to describe statistics characterizing the tool's effectiveness and satisfaction of the tool usage.

Materials and Methods: All 525 BMD examinations individually paired with chest radiographs during the years2022–2023. The AISOT was developed based on deep learning concept on chest radiograph images to predict BMD value. Both BMD observed and predicted values were classified osteoporosis condition by using T-scores of preclinical guidelines. The AISOT demonstrated the accuracy, sensitivity, specificity at 94% (95% CI: 84 -99%), 95% (95% CI: 83-99%) and 90% (95% CI: 56- 99%) respectively. The AISOT model was tested with 195 participants. The research instrument was a questionnaire developed by the researcher including personal health history, observed and predicted BMD values, FRAX tool osteoporosis evaluation and AISOT satisfaction. Z-test was utilized to compare statistics characterized the tool effectiveness.

Results: The AISOT vs the FRAX tool comprised accuracy at 74% (95% CI: 67- 80%) vs 51% (95% CI: 44 -58%) (p < .001); sensitivity at 61% (95% CI: 52-70%) vs 24 % (95% CI: 17-33%) (p = .012); specificity at 93% (95% CI: 85-98%) vs 92% (95% CI: 84-97%) (p > .05); PPV at 94% vs 83% (p = .015); NPV at 61% vs 44% (p = .016). AISOT satisfaction was at very satisfied level (mean = 4.81, SD.= 0.434).

Conclusions: AISOT is effective for screening abnormal BMD in people with osteoporotic risk factors. Future study of various settings will enhance its credibility.

References

National Institute of Arthritis and Musculoskeletal and Skin Diseases. Osteoporosis. [Internet] 2022 [cited 2024 Dec 15]. Available from: https://www.niams.nih.gov/health-topics/osteoporosis#:~:text=Osteoporosis%20is%20a%20bone%20disease,of%20fractures%20(broken%20bones).

Taechakraichana N. Treatment and Prevention of Osteoporosis in Menopause. Siriraj Med J [Internet]. 2005 [cited 2025 Oct 9];57(12):555-6. Available from: https://he02.tci-thaijo.org/index.php/sirirajmedj/article/view/245661

Charoenngam N, Pongchaiyakul C. Current issues in evaluation and management of osteoporosis in Thailand. Osteoporos Sarcopenia. 2023;9(2):53-9.

Pongchaiyakul C, Songpattanasilp T, Taechakraichana N. Burden of Osteoporosis in Thailand. J Med Assoc Thai. 2008;91(2):261-7.

Nuti R, Brandi ML, Checchia G, Di Munno O, Dominguez L, Falaschi P, et al. Guidelines for the management of osteoporosis and fragility fractures. Intern Emerg Med. 2019;14(1):85-102.

Johns Hopkins Medicine. Bone Densitometry. [Internet]. (n.d.) [cited 2024 Jan 10]. Available from: https://www.hopkinsmedicine.org/health/treatment-tests-and-therapies/bone-

Mithal A, Bansal B, Kyer CS, Ebeling P. The Asia-Pacific Regional Audit-Epidemiology, Costs, and Burden of Osteoporosis in India 2013: A report of International Osteoporosis Foundation. Indian J Endocrinol Metab. 2014;18(4):449-54.

Drug and Medical Supply Information Center, Ministry of Public Health. Osteoporosis Screening (W 316). [Internet] 2011 [cited 2024 Jan 20]. Available from: https://dmsic.moph.go.th/index/detail/1385

Sharma S. Drawing insights from COVID-19-infected patients using CT scan images and machine learning techniques: a study on 200 patients. Environ Sci Pollut Res Int. 2020;27(29): 37155-37163.

Whitlock RH, Leslie WD, Shaw J, Rigatto C, Thorlacius L, Komenda P, et al. The Fracture Risk Assessment Tool (FRAX®) predicts fracture risk in patients with chronic kidney disease. Kidney Int. 2019;95(2):447-54.

Webb R. Two proportion Z-test and confidence interval. [Internet]. 2023. [cited 2024 Dec 20]. Available from: https://stats.libretexts.org/Bookshelves/Introductory_Statistics/Mostly_Harmless_Statistics_(Webb)/09%3A_Hypothesis_Tests_and_Confidence_Intervals_for_Two_Populations/9.03%3A_Two_Proportion_Z-Test_and_Confidence_Interval

Liu RW, Ong W, Makmur A, Kumar N, Low XZ, Shuliang G, et al. Application of artificial intelligence methods on osteoporosis classification with radiographs—A Systematic Review. Bioengineering (Basel). 2024;11(5):484.

RadiologyInfo.org. Chest X-ray. Radio Society of North America. [Internet]. 2025. [cited 2024 Dec 12]. Available from: https://www.radiologyinfo.org/en/info/chestrad

Jonathan AC Sterne, Miguel A Hernán, Alexandra McAleenan, Barnaby C Reeves, Julian PT Higgins. Cochrane Handbooks for Systematic Reviews of Interventions: Assessing Risk of Bias in a Non-randomized Study. [Internet]. (n.d.). [cited 2024 Dec 12]. Available from: https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-25

Murad MH, Lin L, Chu H, Hasan B, Alsibai RA, Abbas AS, et al. The association of sensitivity and specificity with disease prevalence: analysis of 6909 studies of diagnostic test accuracy. CMAJ. 2023;195(27):E925-E931.

Caraguel CG, Vanderstichel R. The two-step Fagan's nomogram: ad hoc interpretation of a diagnostic test result without calculation. Evid Based Med. 2013;18(4):125-8.