KI-gestützte Erkennung von Wirbelkörperfrakturen in Routine-CTs: Wie künstliche Intelligenz die Diagnoselücke bei Osteoporose reduzieren könnte

Zusammenfassung

Osteoporotische Wirbelkörperfrakturen (VCFs) sind mit erheblichen gesundheitlichen Folgen verbunden und werden oft übersehen, da sie in zwei Dritteln der Fälle asymptomatisch verlaufen. Eine unerkannte VCF erhöht das Risiko für weitere osteoporotische Frakturen, insbesondere Hüftfrakturen, und stellt eine ernsthafte Belastung für das Gesundheitssystem dar. Trotz etablierter Behandlungsmöglichkeiten bleibt die Versorgungslücke hoch, insbesondere weil viele Wirbelfrakturen nicht erkannt oder nicht dokumentiert werden. Künstliche Intelligenz (KI) bietet eine vielversprechende Möglichkeit, diese diagnostische Lücke zu schließen. Verschiedene KI-gestützte Systeme ermöglichen die automatische Erkennung klinisch relevanter VCFs auf Routine-CTs, wodurch Frakturen frühzeitiger identifiziert und PatientInnen einer gezielten osteologischen Abklärung zugeführt werden können. Moderne Deep-Learning-Modelle analysieren CT-Daten präzise, bieten eine standardisierte und nutzerunabhängige Befundung und reduzieren den zusätzlichen Arbeitsaufwand für RadiologInnen. Mehrere auf dem Markt verfügbare KI-Lösungen haben bereits eine hohe diagnostische Genauigkeit gezeigt, mit Sensitivitäts- und Spezifitätswerten im Bereich von 80–95% bzw. 93–99%. Am Beispiel der CE-zertifizierten Software IB Lab FLAMINGO wird gezeigt, wie Künstliche Intelligenz dazu beitragen kann, klinisch relevante VCFs opportunistisch zu identifizieren. Durch eine frühzeitige Identifikation betroffener PatientInnen können Therapiemaßnahmen gezielt eingeleitet und die Versorgung optimiert werden. Langfristig könnte die Integration von KI in die Frakturdiagnostik dazu beitragen, die Behandlungslücke in der Osteoporoseversorgung signifikant zu reduzieren.

Abstract

Osteoporotic vertebral compression fractures (VCFs) are associated with significant health consequences and often go unnoticed, as two-thirds of cases are asymptomatic. An undiagnosed VCF increases the risk of subsequent osteoporotic fractures, particularly hip fractures, and poses a serious burden on healthcare systems. Despite established treatment options, a significant care gap remains, especially because many vertebral fractures are either not detected or not reported. Artificial intelligence (AI) offers a promising approach to closing this diagnostic gap. Various AI-powered systems enable the automated detection of clinically relevant VCFs on routine CT scans, allowing for earlier fracture identification and timely osteological assessment. Advanced deep-learning models analyze CT data with high precision, provide standardized and objective reporting, and minimize additional workload for radiologists. Several AI solutions currently available on the market have demonstrated high diagnostic accuracy, with sensitivity and specificity ranging from 80–95% and 93–99%, respectively.Using the example of the CE-certified IB Lab FLAMINGO software, it will be shown how artificial intelligence can help to opportunistically identify clinically relevant VCFs. By facilitating the timely identification of at-risk patients, AI-driven solutions can enable earlier therapeutic intervention and improve overall patient management. In the long term, integrating AI into fracture diagnostics could significantly contribute to reducing the osteoporosis treatment gap.

Schlüsselwörter

Osteoporose - CT - Frakturerkennung - KI - Wirbelkörperfraktur

Keywords

Osteoporosis - CT - fracture detection - AI - vertebral fracture

Publication History

Received: 19 March 2025

Accepted: 24 July 2025

Article published online:
19 August 2025

• Literatur

• 1 Sanfélix-Genovés J, Hurtado I, Sanfélix-Gimeno G. et al. Impact of osteoporosis and vertebral fractures on quality-of-life. a population-based study in Valencia, Spain (The FRAVO Study). Health Qual Life Outcomes 2011; 9: 20
• 2 Lems WF, Paccou J, Zhang J. et al. Vertebral fracture: epidemiology, impact and use of DXA vertebral fracture assessment in fracture liaison services. Osteoporos Int 2021; 32: 399-411
• 3 Kanis JA, Cooper C, Rizzoli R. et al. European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 2019; 30: 3-44
• 4 Kanis JA, Norton N, Harvey NC. et al. SCOPE 2021: a new scorecard for osteoporosis in Europe. Arch Osteoporos 2021; 16: 82
• 5 Noguchi T, Yamashita K, Kamei R. et al. Current status and challenges of percutaneous vertebroplasty (PVP). Jpn J Radiol 2023; 41: 1-13
• 6 Kanis JA, Borgstrom F, De Laet C. et al. Assessment of fracture risk. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2005; 16: 581-589
• 7 Skjødt MK, Nicolaes J, Smith CD. et al. Fracture Risk in Men and Women With Vertebral Fractures Identified Opportunistically on Routine Computed Tomography Scans and Not Treated for Osteoporosis: An Observational Cohort Study. JBMR Plus 2023; n/a e10736
• 8 Schulz C, König H-H, Rapp K. et al. Analysis of mortality after hip fracture on patient, hospital, and regional level in Germany. Osteoporos Int 2020; 31: 897-904
• 9 Autier P, Haentjens P, Bentin J. et al. Costs induced by hip fractures: a prospective controlled study in Belgium. Belgian Hip Fracture Study Group. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2000; 11: 373-380
• 10 DVO e.V. Prophylaxe, Diagnostik und Therapie der Osteoporose. Leitlinie des Dachverbands der Deutschsprachigen Wissenschaftlichen Osteologischen Gesellschaft e.V. Im Internet: https://leitlinien.dv-osteologie.org/wp-content/uploads/2024/02/DVO-Leitlinie-zur-Diagnostik-und-Therapie-der-Osteoporose-Version-2.1.-2023-002.pdf
• 11 IOF International Osteoporosis Foundation. EPIDEMIOLOGY OF OSTEOPOROSIS AND FRAGILITY FRACTURES. Im Internet: https://www.osteoporosis.foundation/facts-statistics Stand: 18.07.2024
• 12 Harvey NCW, McCloskey EV, Mitchell PJ. et al. Mind the (treatment) gap: a global perspective on current and future strategies for prevention of fragility fractures. Osteoporos Int 2017; 28: 1507-1529
• 13 Kanis JA, Cooper C, Rizzoli R. et al. Identification and management of patients at increased risk of osteoporotic fracture: outcomes of an ESCEO expert consensus meeting. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2017; 28: 2023-2034
• 14 Kocijan R, Haschka J, Kraus DA. et al. Perspectives on Fracture Liaison Service in Austria: clinical and economic considerations. Front Endocrinol 2024; 15: 1349579
• 15 Williams AL, Al-Busaidi A, Sparrow PJ. et al. Under-reporting of osteoporotic vertebral fractures on computed tomography. Eur J Radiol 2009; 69: 179-183
• 16 Carberry GA, Pooler BD, Binkley N. et al. Unreported vertebral body compression fractures at abdominal multidetector CT. Radiology 2013; 268: 120-126
• 17 Löffler MT, Kallweit M, Niederreiter E. et al. Epidemiology and reporting of osteoporotic vertebral fractures in patients with long-term hospital records based on routine clinical CT imaging. Osteoporos Int 2022; 33: 685-694
• 18 Bartalena T. Incidental vertebral compression fractures in imaging studies: Lessons not learned by radiologists. World J Radiol 2010; 2: 399
• 19 Bartalena T, Giannelli G, Rinaldi MF. et al. Prevalence of thoracolumbar vertebral fractures on multidetector CT. Eur J Radiol 2009; 69: 555-559
• 20 Lai Y. A Comparison of Traditional Machine Learning and Deep Learning in Image Recognition. J Phys Conf Ser 2019; 1314: 012148
• 21 Aggarwal V, Maslen C, Abel RL. et al. Opportunistic diagnosis of osteoporosis, fragile bone strength and vertebral fractures from routine CT scans; a review of approved technology systems and pathways to implementation. Ther Adv Musculoskelet Dis 2021; 13 1759720X211024029
• 22 Engelke K, Chaudry O, Bartenschlager S. Opportunistic Screening Techniques for Analysis of CT Scans. Curr Osteoporos Rep 2023; 21: 65-76
• 23 Genant HK, Wu CY, van Kuijk C. et al. Vertebral fracture assessment using a semiquantitative technique. J Bone Miner Res Off J Am Soc Bone Miner Res 1993; 8: 1137-1148
• 24 Löffler MT, Jacob A, Scharr A. et al. Automatic opportunistic osteoporosis screening in routine CT: improved prediction of patients with prevalent vertebral fractures compared to DXA. Eur Radiol 2021; 31: 6069-6077
• 25 Lentle BC, Prior JC. Osteoporotic vertebral fracture (OVF): diagnosis requires an informed observer. Osteoporos Int 2022; 33: 1409-1410
• 26 Namireddy SR, Gill SS, Peerbhai A. et al. Artificial intelligence in risk prediction and diagnosis of vertebral fractures. Sci Rep 2024; 14: 30560
• 27 Spangeus A, Baldimtsi E, Lindblom M. et al. The VERTAIDO Project (Vertebral Fracture AI Detection for Better Osteoporosis Care): Implementation of AI-Screnning in Clinical Routine - First Month Data. Poster Presentation WCO_IOF_ESCEO 2025, Abstract book (p. 266): https://www.wco-iof-esceo.org/download/2025/abstract-book
• 28 Jiang G, Eastell R, Barrington NA. et al. Comparison of methods for the visual identification of prevalent vertebral fracture in osteoporosis. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2004; 15: 887-896
• 29 Wiklund P, Buchebner D, Geijer M. Vertebral compression fractures at abdominal CT: underdiagnosis, undertreatment, and evaluation of an AI algorithm. J Bone Miner Res Off J Am Soc Bone Miner Res 2024; 39: 1113-1119
• 30 Rueckel J, Sperl JI, Kaestle S. et al. Reduction of missed thoracic findings in emergency whole-body computed tomography using artificial intelligence assistance. Quant Imaging Med Surg 2021; 11: 2486-2498
• 31 FDA. 510k Summary Nanox HealthOST. 2022;
• 32 Dai M, Tiu B-C, Schlossman J. et al. Validation of a Deep Learning Tool for Detection of Incidental Vertebral Compression Fractures. J Comput Assist Tomogr 2025;
• 33 Guenoun MQP Champsaur, A Ayobi, C Charlotte, S Quenet, J Kiewsky, M Mahfoud, C Avare, D. Performance evaluation of an AI-based application for opportunistic screening of thoraco-lumbar vertebral compression fractures. ECR 2024 EPOS 2024; Im Internet: https://epos.myesr.org/poster/esr/ecr2024/C-11881 Stand: 23.07.2024
• 34 Spångeus A, Bjerner T, Lindblom M. et al. Breaking the silence: AI’s contribution to detecting vertebral fractures in opportunistic CT scans in the elderly—a validation study. Arch Osteoporos 2025; 20: 42
• 35 Kraus DA, Zandieh S, Haschka J. et al. Opportunistic Vertebral Fracture Detection from Computed Tomography Studies by Artificial Intelligence - preliminary results: Annual Meeting of the American-Society-for-Bone-and-Mineral-Research (ASBMR). J BONE Miner Res 2023; 38: 344-344
• 36 Sollmann N, Löffler MT, El Husseini M. et al. Automated Opportunistic Osteoporosis Screening in Routine Computed Tomography of the Spine: Comparison With Dedicated Quantitative CT. J Bone Miner Res Off J Am Soc Bone Miner Res 2022; 37: 1287-1296
• 37 REGULATION (EU) 2017/745 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL. Im Internet: https://eur-lex.europa.eu/eli/reg/2017/745/oj/eng
• 38 Nicolaes J, Liu Y, Zhao Y. et al. External validation of a convolutional neural network algorithm for opportunistically detecting vertebral fractures in routine CT scans. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA 2023;
• 39 Omoumi P, Ducarouge A, Tournier A. et al. To buy or not to buy-evaluating commercial AI solutions in radiology (the ECLAIR guidelines). Eur Radiol 2021; 31: 3786-3796