Radiologische Diagnostik und Prognostik von COVID-19: Einsatz von künstlicher Intelligenz und Zusammenhang mit muskulo-skelettaler Bildgebung

Zusammenfassung

Die COVID-19 Pandemie hat die Radiologie, wie viele andere Bereiche, vor völlig neue Herausforderungen gestellt. Radiologische Bildgebung spielt im Verbund mit Laboruntersuchungen und klinischen Daten eine wichtige Rolle bei der Diagnose von COVID-19. Anhand einer spezifischen Analyse der Lungenläsionen erlaubt sie auch Einschätzungen des Risikos schwerer Verläufe, wenngleich die Größenordnung des Informationsgewinns über biologisch klinische Daten hinaus im Einzelfall unterschiedlich und Gegenstand aktueller Forschung ist. Osteoporose-bedingte Frakturen stellen in diesem Zusammenhang möglicherweise einen unabhängigen Risikofaktor für schwere Verläufe dar. Die Pandemie hat aber auch neue Perspektiven eröffnet, insbesondere sind im Bereich der Bildgebung neue technologische Entwicklungen mit Nachdruck vorangetrieben worden. So arbeiten alle Universitätsradiologien am Projekt Radiological Cooperative Network (RACOON) zusammen, wobei auf der Basis strukturierter Befunde die Daten zusammengeführt und optional mit Methoden der künstlichen Intelligenz (KI) analysiert werden. Eine Zusammenführung mit KI-Methoden zur Frakturerkennungen bietet Perspektiven, Frakturinformationen automatisch zu gewinnen und in Risiko-Scores für schweren Verlauf mit einzubinden. Die neuen Strukturen und Methoden, die während der Pandemie entwickelt wurden, lassen sich auf andere Anwendungsbereiche wie die muskulo-skelettale Bildgebung übertragen und können so zu erheblichen technologischen Fortschritten in der radiologischen Diagnostik und Prognostik führen.

Abstract

The COVID-19 pandemic has confronted radiology like many other fields with completely new challenges. Radiological imaging, in conjunction with laboratory tests and clinical data, plays an important role in the diagnosis of COVID-19. Based on specific analyses of lung lesions it also permits estimation of the risk of severe disease outcome – still, the magnitude of information gain beyond biological clinical data varies in individual cases and is the subject of current research. In this context osteoporosis related fractures may represent an independent risk factor for severe disease outcome. However, the pandemic has also opened up new perspectives, and in particular new technological developments have been vigorously pursued in the field of imaging. For example, all university radiology departments are collaborating on the Radiological Cooperative Network (RACOON) project, where data are merged based on structured reporting and those can optionally be analyzed using artificial intelligence (AI) methods. Merging these approaches with AI methods for fracture detection offers prospects for automatically obtaining fracture information and incorporating it into risk scores for severe disease outcome. The new structures and methods developed during the pandemic can be transferred to other application fields such as musculo-skeletal imaging and may thus bring about substantial technologically advances for radiological diagnostics and prognostics.

Schlüsselwörter

Radiologie - COVID-19 - künstliche Intelligenz - Frakturen

Key words

radiology - COVID-19 - artificial intelligence - fracture

Publikationsverlauf

Eingereicht: 08. November 2021

Angenommen nach Revision: 07. Dezember 2021

Artikel online veröffentlicht:
21. Februar 2022

• Literatur

• 1 di Filippo L, Formenti AM, Doga M. et al. Radiological Thoracic Vertebral Fractures are Highly Prevalent in COVID-19 and Predict Disease Outcomes. J Clin Endocrinol Metab 2021; 106: e602-e614
• 2 Battisti S, Napoli N, Pedone C. et al. Vertebral fractures and mortality risk in hospitalised patients during the COVID-19 pandemic emergency. Endocrine 2021; DOI: 10.1007/s12020-021-02872-1.
• 3 Schlaich C, Minne HW, Bruckner T. et al. Reduced Pulmonary Function in Patients with Spinal Osteoporotic Fractures. Osteoporos Int 1998; 8: 261-267 DOI: 10.1007/s001980050063.
• 4 Minne HW, Leidig G, Wüster C. et al. A newly developed spine deformity index (SDI) to quantitate vertebral crush fractures in patients with osteoporosis. Bone Miner 1988; 3: 335-349
• 5 Leech JA, Dulberg C, Kellie S. et al. Relationship of Lung Function to Severity of Osteoporosis in Women. American Review of Respiratory Disease 1990; DOI: 10.1164/ajrccm/141.1.68.
• 6 Harrison RA, Siminoski K, Vethanayagam D. et al. Osteoporosis-Related Kyphosis and Impairments in Pulmonary Function: A Systematic Review. Journal of Bone and Mineral Research 2007; 22: 447-457 DOI: 10.1359/jbmr.061202.
• 7 Morseth B, Melbye H, Waterloo S. et al. Cross-sectional associations between prevalent vertebral fracture and pulmonary function in the sixth Tromsø study. BMC Geriatr 2013; 13: 116 DOI: 10.1186/1471-2318-13-116.
• 8 Krege JH, Kendler D, Krohn K. et al. Relationship Between Vertebral Fracture Burden, Height Loss, and Pulmonary Function in Postmenopausal Women With Osteoporosis. Journal of Clinical Densitometry 2015; 18: 506-511 DOI: 10.1016/j.jocd.2015.02.004.
• 9 Crans GG, Genant HK, Krege JH. Prognostic utility of a semiquantitative spinal deformity index. Bone 2005; 37: 175-179 DOI: 10.1016/j.bone.2005.04.003.
• 10 Isensee F, Jaeger PF, Kohl SAA. et al. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat Methods 2021; 18: 203-211 DOI: 10.1038/s41592-020-01008-z.
• 11 Images Download - 2019 Novel Coronavirus Resource (2019nCoVR). Im Internet: http://ncov-ai.big.ac.cn/download?lang=en Stand: 03.11.2021
• 12 Yilmaz EB, Bürger C, Fricke T. et al. Automated Deep Learning-Based Detection of Osteoporotic Fractures in CT Images. In: Lian C, Cao X, Rekik I, Xu X, Yan P, Hrsg. Machine Learning in Medical Imaging – 12th International Workshop, MLMI 2021, Held in Conjunction with MICCAI 2021, Strasbourg, France, September 27, 2021, Proceedings. Springer; 2021: 376-385 DOI: 10.1007/978-3-030-87589-3_39
• 13 Buerger C, von Berg J, Franz A. et al. Combining deep learning and model-based segmentation for labeled spine CT segmentation. In: Landman BA, Išgum I, Hrsg. Medical Imaging 2020: Image Processing. Houston, United States: SPIE; 2020: 47 DOI: 10.1117/12.2549485
• 14 Yilmaz E. Towards an automatic system for assessment of vertebral fractures in CT images. 2020
• 15 Glüer C-C, Krause M, Museyko O. et al. New horizons for the in vivo assessment of major aspects of bone quality: Microstructure and material properties assessed by Quantitative Computed Tomography and Quantitative Ultrasound methods developed by the BioAsset consortium. Osteologie 2013; 22: 223-233 DOI: 10.1055/s-0038-1630120.
• 16 Kottlors J, Große Hokamp N, Fervers P. et al. Early extrapulmonary prognostic features in chest computed tomography in COVID-19 pneumonia: Bone mineral density is a relevant predictor for the clinical outcome - A multicenter feasibility study. Bone 2021; 144: 115790 DOI: 10.1016/j.bone.2020.115790.
• 17 Sideris GA, Nikolakea M, Karanikola A-E. et al. Imaging in the COVID-19 era: Lessons learned during a pandemic. WJR 2021; 13: 192-222 DOI: 10.4329/wjr.v13.i6.192.
• 18 Vogel-Claussen J, Ley-Zaporozhan J, Agarwal P. et al. Recommendations of the Thoracic Imaging Section of the German Radiological Society for clinical application of chest imaging and structured CT reporting in the COVID-19 pandemic. Röfo 2020; 192: 633-640 DOI: 10.1055/a-1174-8378.
• 19 Kluge S, Janssens U, Welte T. et al. S3-Leitlinie – Empfehlungen zur stationären Therapie von Patienten mit COVID-19 2021; 89
• 20 ACR Recommendations for the use of Chest Radiography and Computed Tomography (CT) for Suspected COVID-19 Infection | American College of Radiology. Im Internet: https://www.acr.org/Advocacy-and-Economics/ACR-Position-Statements/Recommendations-for-Chest-Radiography-and-CT-for-Suspected-COVID19-Infection Stand: 08.11.2021
• 21 Antoch G, Urbach H, Mentzel H-J. et al. SARS-CoV-2/COVID-19: Empfehlungen für die Radiologische Versorgung – Eine Stellungnahme der Deutschen Röntgengesellschaft (DRG), der Deutschen Gesellschaft für Neuroradiologie (DGNR), der Gesellschaft für Pädiatrische Radiologie (GPR), der Deutschen Gesellschaft für Interventionelle Radiologie (DeGIR), des Berufsverbands der Neuroradiologen (BDNR) und des Berufsverbands der Radiologen (BDR). Rofo 2020; 192: 418-421 DOI: 10.1055/a-1149-3625.
• 22 Kovács A, Palásti P, Veréb D. et al. The sensitivity and specificity of chest CT in the diagnosis of COVID-19. Eur Radiol 2021; 31: 2819-2824 DOI: 10.1007/s00330-020-07347-x.
• 23 Liang W, Liang H, Ou L. et al. Development and Validation of a Clinical Risk Score to Predict the Occurrence of Critical Illness in Hospitalized Patients With COVID-19. JAMA Intern Med 2020; DOI: 10.1001/jamainternmed.2020.2033.
• 24 Maroldi R, Rondi P, Agazzi GM. et al. Which role for chest x-ray score in predicting the outcome in COVID-19 pneumonia?. Eur Radiol 2021; 31: 4016-4022 DOI: 10.1007/s00330-020-07504-2.
• 25 Gong Y, Guan L, Jin Z. et al. Effects of methylprednisolone use on viral genomic nucleic acid negative conversion and CT imaging lesion absorption in COVID-19 patients under 50 years old. Journal of Medical Virology 2020; 92: 2551-2555 DOI: 10.1002/jmv.26052.
• 26 Salg GA, Ganten M-K, Bucher AM. et al. A reporting and analysis framework for structured evaluation of COVID-19 clinical and imaging data. npj Digit Med 2021; 4: 69 DOI: 10.1038/s41746-021-00439-y.
• 27 urbanstudio.de. Information der AG Thoraxdiagnostik der DRG zur klinischen Anwendung und strukturierten Befundung der Computertomographie (CT) bei COVID-19 Epidemie | AG Thoraxdiagnostik. Im Internet: https://www.ag-thorax.drg.de Stand: 08.11.2021
• 28 Sistrom CL, Honeyman-Buck J. Free text versus structured format: information transfer efficiency of radiology reports. AJR Am J Roentgenol 2005; 185: 804-812 DOI: 10.2214/ajr.185.3.01850804.
• 29 Willemink MJ, Koszek WA, Hardell C. et al. Preparing Medical Imaging Data for Machine Learning. Radiology 2020; 295: 4-15 DOI: 10.1148/radiol.2020192224.
• 30 Yamashita R, Nishio M, Do RKG. et al. Convolutional neural networks: an overview and application in radiology. Insights Imaging 2018; 9: 611-629 DOI: 10.1007/s13244-018-0639-9.
• 31 Zhang K, Liu X, Shen J. et al. Clinically Applicable AI System for Accurate Diagnosis, Quantitative Measurements, and Prognosis of COVID-19 Pneumonia Using Computed Tomography. Cell 2020; 181: 1423-1433.e11 DOI: 10.1016/j.cell.2020.04.045.
• 32 Wang H, Wei R, Rao G. et al. Characteristic CT findings distinguishing 2019 novel coronavirus disease (COVID-19) from influenza pneumonia. Eur Radiol 2020 DOI: 10.1007/s00330-020-06880-z.
• 33 Roberts M, Driggs D, Thorpe M. et al. Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans. Nature Machine Intelligence 2021; 3: 199-217 DOI: 10.1038/s42256-021-00307-0.
• 34 Scherer J, Nolden M, Kleesiek J. et al. Joint Imaging Platform for Federated Clinical Data Analytics. JCO Clinical Cancer Informatics 2020; 1027-1038 DOI: 10.1200/CCI.20.00045.
• 35 Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 1996; 312: 1254-1259
• 36 Hans D, Goertzen AL, Krieg M-A. et al. Bone microarchitecture assessed by TBS predicts osteoporotic fractures independent of bone density: the Manitoba study. J Bone Miner Res 2011; 26: 2762-2769 DOI: 10.1002/jbmr.499.
• 37 Iki M, Winzenrieth R, Tamaki J. et al. Predictive ability of novel volumetric and geometric indices derived from dual-energy X-ray absorptiometric images of the proximal femur for hip fracture compared with conventional areal bone mineral density: The Japanese Population-based Osteoporosis (JPOS) Cohort Study. Osteoporos Int 2021; DOI: 10.1007/s00198-021-06013-2.
• 38 Kopperdahl DL, Aspelund T, Hoffmann PF. et al. Assessment of Incident Spine and Hip Fractures in Women and Men using Finite Element Analysis of CT Scans. J Bone Miner Res 2014; 29: 570-580 DOI: 10.1002/jbmr.2069.
• 39 Bouxsein ML, Zysset P, Glüer C-C. et al. Perspectives on the non-invasive evaluation of femoral strength in the assessment of hip fracture risk. Osteoporos Int 2020; DOI: 10.1007/s00198-019-05195-0.
• 40 Shrimpton PC, Wall BF, Jones DG. et al. A national survey of doses to patients undergoing a selection of routine x-ray examinations in English hospitals (NRPB-R–200). United Kingdom: 1986
• 41 Museyko O, Heinemann A, Krause M. et al. A low-radiation exposure protocol for 3D QCT of the spine. Osteoporos Int 2014; 25: 983-992 DOI: 10.1007/s00198-013-2544-x.
• 42 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 DOI: 10.1007/s00330-020-07655-2.
• 43 Kolanu N, Silverstone EJ, Ho BH. et al. Clinical Utility of Computer-Aided Diagnosis of Vertebral Fractures From Computed Tomography Images. Journal of Bone and Mineral Research 2020; 35: 2307-2312 DOI: 10.1002/jbmr.4146.
• 44 Smets J, Shevroja E, Hügle T. et al. Machine Learning Solutions for Osteoporosis—A Review. J Bone Miner Res 2021; 36: 833-851 DOI: 10.1002/jbmr.4292.
• 45 Damm T, Schmarje L, Koser N. et al. Artificial intelligence-driven hip fracture prediction based on pelvic radiographs exceeds performance of DXA: the “Study of Osteoporotic Fractures” (SOF) Annual Meeting American Society for Bone and Mineral Research, San Diego,. 2021