Back to the list
Congress: ECR25
Poster Number: C-16938
Type: Poster: EPOS Radiologist (scientific)
Authorblock: M. Woisetschlager1, T. Bjerner1, M. Lindblom1, C. Götz2, A. Hummer2, C. Salzlechner2, A. Spångeus1; 1Linköping/SE, 2Vienna/AT
Disclosures:
Mischa Woisetschlager: Nothing to disclose
Tomas Bjerner: Nothing to disclose
Maria Lindblom: Nothing to disclose
Christoph Götz: Employee: IB lab GmbH
Allan Hummer: Employee: IB lab GmbH
Christoph Salzlechner: Employee: IB lab GmbH
Anna Spångeus: Advisory Board: UCB Consultant: Giddeon Richter Speaker: Amgen, Tromp Medical
Keywords: Abdomen, Artificial Intelligence, Computer applications, CT, Computer Applications-General, Osteoporosis
Purpose Osteoporotic vertebral fractures (VF) are associated with significant morbidity, mortality, and an increased risk of subsequent fractures. Today, there are effective treatment alternatives available that significantly lower the risk of subsequent fractures. Despite this, over two-thirds of VFs remain undetected by radiologists, underscoring the necessity for enhanced detection methods. Opportunistic screening for VFs is a high priority in the field of osteoporosis. Recently, AI-based solutions for automating vertebral fracture detection in CT examinations have emerged, which could be highly valuable...
Read more Methods and materials Cohort: The study utilized 246 existing CT scans originally conducted for purposes unrelated to vertebral assessment but which included the thoracic and/or lumbar spine from a prior study on in-hospital falls. The cohort had a mean age of 84 years (range 62-103) with 42% being female. The CT scans included both women and men, thoracic or abdominal examinations, and were performed with various CT protocols, including bone and non-bone kernel, as well as different contrast phases. Method: AI analysis was performed...
Read more Results Entire cohort: Compared to the ground truth, the AI demonstrated high accuracy (0.93), sensitivity (0.86), and specificity (0.99) in detecting moderate to severe VFs.Subgroup analysis: Subgroup analysis indicated accuracy ranging from 0.88 to 0.96, with higher accuracy observed in females compared to males (0.96 vs. 0.89, p=0.03) and in scans using non-bone versus bone kernel protocols (0.96 vs. 0.88, p=0.02). No significant differences were found concerning age, contrast agent use, or anatomic region. [fig 3]
Read more Conclusion This study validates the high accuracy, sensitivity, and specificity of an AI algorithm in detecting moderate to severe vertebral fractures in a geriatric cohort. The AI demonstrated particularly high accuracy in females and in scans using non-bone kernel protocols. These findings highlight the potential of AI-based solutions to enhance opportunistic screening for vertebral fractures, addressing a critical gap in osteoporosis management.
Read more References Banefelt J, Åkesson KE, Spångéus A, et al. (2019) Risk of imminent fracture following a previous fracture in a Swedish database study. Osteoporosis International. 2019;30(3):601-609. doi:10.1007/s00198-019-04852-8Skjødt MK, Nicolaes J, Smith CD, et al. (2024) Opportunistically identifiable vertebral fractures on routine radiological imaging predict mortality: observational cohort study. Osteoporosis International. 2024;35(4):691-703. doi:10.1007/s00198-024-07017-4Toth E, Banefelt J, Åkesson K et al. (2020) History of Previous Fracture and Imminent Fracture Risk in Swedish Women Aged 55-90 Years Presenting with a Fragility Fracture. Journal of...
Read more
GALLERY