Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Mar 25;20(1):42.
doi: 10.1007/s11657-025-01524-5.

Breaking the silence: AI's contribution to detecting vertebral fractures in opportunistic CT scans in the elderly-a validation study

Anna Spångeus  1   2   3   4 , Tomas Bjerner  5   6   7 , Maria Lindblom  7 , Christoph Götz  8 , Allan Hummer  8 , Christoph Salzlechner  8 , Mischa Woisetschläger  5   6   7
Affiliations

Breaking the silence: AI's contribution to detecting vertebral fractures in opportunistic CT scans in the elderly-a validation study

Anna Spångeus et al. Arch Osteoporos. .

Abstract

Vertebral fractures frequently go undetected in clinical practice. AI-assisted detection on CT scans demonstrates considerable promise, with a sensitivity of 86% and a specificity of 99%. The performance varied based on sex, and CT kernel, showing superior results in females and in scans using non-bone kernel protocols.

Purpose: Vertebral fractures (VFs) are highly underdiagnosed, necessitating the development of new identification methods for opportunistic screening in computed tomography (CT) scans. This study validated an AI algorithm (ImageBiopsy Lab [IBL], FLAMINGO) for detecting VFs in a geriatric cohort, with various subgroup analyses including different CT protocols.

Methods: The performance of the AI in detecting VFs was compared to assessments by two experienced radiologists. A total of 246 thoracic or abdominal CT scans, primarily conducted for purposes other than skeletal examination, were included in the study.

Results: The patients had a mean age of 84 years (range 62 to 103), with 42% being female. The AI demonstrated high accuracy (0.93), sensitivity (0.86), and specificity (0.99) in detecting moderate to severe VFs. Subgroup analysis revealed accuracy ranging from 0.88 to 0.96, with higher accuracy in females compared to males (0.96 vs. 0.89, p = 0.03) and in scans performed with non-bone kernel versus bone kernel protocols (0.96 vs. 0.88, p = 0.02). No significant differences were found for age, contrast phase, or spinal region.

Conclusion: The results indicate that the AI algorithm exhibits high performance in a geriatric setting. If effectively integrated with a fracture liaison service, this could enhance VF detection considerable in the future.

Keywords: Artificial intelligence; Computer tomography; Geriatric; Opportunistic screening; Osteoporosis; Vertebral fracture.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics approval: The study was approved by the Regional Research Ethics Committee of the Faculty of Health Sciences, Sweden (2021–01425 and 2022–02053-02). Informed consent: In accordance with ethical approval, informed consent was not needed for this retrospective study. Permission to reproduce material from other sources: Not applicable. Clinical trial registration: Not applicable. Conflict of interest: Author AS has received fees for lecturer from UCB, AMGEN, and Tromp Medical and for advisory board from UCB. Authors MW, TB, and ML have no conflicts of interest in this work. CG, AH, and CS are employees of IB Lab GmbH.

Figures

Fig. 1
Fig. 1
Images showing examples of different contrast phases and kernels. A Datasets without (left) and with (right) contrast agent visualizing the uptake of contrast agent in the aorta, kidneys, liver, pancreas and spleen, and to a lesser extend in the trabecular portion of the vertebrae. B Datasets with bone kernel (left) and non-bone kernel (right) illustrating the differences in image noise and edge sharpness
Fig. 2
Fig. 2
Result from IB Lab FLAMINGO illustrating which vertebrae were analyzed (here T10–L5) and which vertebrae were found fractured (here L1). Only vertebral fractures with a Genant grade of 2 or 3 were reported
See this image and copyright information in PMC

References

    1. Lauppe R, Åkesson KE, Ljunggren Ö et al (2019) Differing impact of clinical factors on the risk of fracture in younger and older women in the general population and an osteoporosis clinic population. Arch Osteoporos 14:45. 10.1007/s11657-019-0592-3 - PubMed
    1. Banefelt J, Åkesson KE, Spångéus A et al (2019) Risk of imminent fracture following a previous fracture in a Swedish database study. Osteoporos Int 30:601–609. 10.1007/s00198-019-04852-8 - PMC - PubMed
    1. Skjødt MK, Nicolaes J, Smith CD et al (2024) Opportunistically identifiable vertebral fractures on routine radiological imaging predict mortality: observational cohort study. Osteoporos Int 35:691–703. 10.1007/s00198-024-07017-4 - PMC - PubMed
    1. Toth E, Banefelt J, Åkesson K, Spångeus A, Orsäter G, Libanati C (2020) History of previous fracture and imminent fracture risk in Swedish women aged 55–90 years presenting with a fragility fracture. J Bone Miner Res 35:861–868. 10.1002/jbmr.3953 - PMC - PubMed
    1. Black DM, Delmas PD, Eastell R et al (2007) Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis. N\ Engl J Med. 10.1056/NEJMoa067312. (n engl j med 35618) - PubMed

Publication types