How artificial intelligence improves radiological interpretation in suspected pulmonary embolism.
Artificial intelligence
Computed tomography angiography
Predictive value of tests
Pulmonary embolism
Sensitivity and specificity
Journal
European radiology
ISSN: 1432-1084
Titre abrégé: Eur Radiol
Pays: Germany
ID NLM: 9114774
Informations de publication
Date de publication:
Sep 2022
Sep 2022
Historique:
received:
29
12
2021
accepted:
04
02
2022
revised:
29
12
2021
pubmed:
23
3
2022
medline:
19
8
2022
entrez:
22
3
2022
Statut:
ppublish
Résumé
To evaluate and compare the diagnostic performances of a commercialized artificial intelligence (AI) algorithm for diagnosing pulmonary embolism (PE) on CT pulmonary angiogram (CTPA) with those of emergency radiologists in routine clinical practice. This was an IRB-approved retrospective multicentric study including patients with suspected PE from September to December 2019 (i.e., during a preliminary evaluation period of an approved AI algorithm). CTPA quality and conclusions by emergency radiologists were retrieved from radiological reports. The gold standard was a retrospective review of CTPA, radiological and clinical reports, AI outputs, and patient outcomes. Diagnostic performance metrics for AI and radiologists were assessed in the entire cohort and depending on CTPA quality. Overall, 1202 patients were included (median age: 66.2 years). PE prevalence was 15.8% (190/1202). The AI algorithm detected 219 suspicious PEs, of which 176 were true PEs, including 19 true PEs missed by radiologists. In the cohort, the highest sensitivity and negative predictive values (NPVs) were obtained with AI (92.6% versus 90% and 98.6% versus 98.1%, respectively), while the highest specificity and positive predictive value (PPV) were found with radiologists (99.1% versus 95.8% and 95% versus 80.4%, respectively). Accuracy, specificity, and PPV were significantly higher for radiologists except in subcohorts with poor-to-average injection quality. Radiologists positively evaluated the AI algorithm to improve their diagnostic comfort (55/79 [69.6%]). Instead of replacing radiologists, AI for PE detection appears to be a safety net in emergency radiology practice due to high sensitivity and NPV, thereby increasing the self-confidence of radiologists. • Both the AI algorithm and emergency radiologists showed excellent performance in diagnosing PE on CTPA (sensitivity and specificity ≥ 90%; accuracy ≥ 95%). • The AI algorithm for PE detection can help increase the sensitivity and NPV of emergency radiologists in clinical practice, especially in cases of poor-to-moderate injection quality. • Emergency radiologists recommended the use of AI for PE detection in satisfaction surveys to increase their confidence and comfort in their final diagnosis.
Identifiants
pubmed: 35316363
doi: 10.1007/s00330-022-08645-2
pii: 10.1007/s00330-022-08645-2
pmc: PMC8938594
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
5831-5842Informations de copyright
© 2022. The Author(s), under exclusive licence to European Society of Radiology.
Références
Smith SB, Geske JB, Maguire JM et al (2010) Early anticoagulation is associated with reduced mortality for acute pulmonary embolism. Chest 137:1382–1390
doi: 10.1378/chest.09-0959
Konstantinides SV, Meyer G, Becattini C et al (2019) 2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism developed in collaboration with the European Respiratory Society (ERS). Eur Heart J 41:543–603
Essien EO, Rali P, Mathai SC (2019) Pulmonary Embolism. Med Clin North Am 103:549–564
doi: 10.1016/j.mcna.2018.12.013
Suhail Akhter M, Hamali HA, Mobarki AA et al (2021) Clinical medicine SARS-CoV-2 infection: modulator of pulmonary embolism paradigm. J Clin Med 10:1064
doi: 10.3390/jcm10051064
Barragán-Montero A, Javaid U, Valdés G et al (2021) Artificial intelligence and machine learning for medical imaging: a technology review. Phys Medica 83:242–256
doi: 10.1016/j.ejmp.2021.04.016
Prevedello LM, Little KJ, Qian S, White RD (2017) Artificial intelligence in imaging 1. Radiology 285:923–931
doi: 10.1148/radiol.2017162664
Lee JY, Kim JS, Kim TY, Kim YS (2020) Detection and classification of intracranial haemorrhage on CT images using a novel deep learning algorithm. Sci Rep 10:20546 1–7. https://doi.org/10.1038/s41598-020-77441-z
doi: 10.1038/s41598-020-77441-z
pubmed: 33239711
pmcid: 7689498
Nagel S, Sinha D, Day D et al (2016) e-ASPECTS software is non-inferior to neuroradiologists in applying the ASPECT score to computed tomography scans of acute ischemic stroke patients. Int J Stroke 0:1–8. https://doi.org/10.1177/1747493016681020
doi: 10.1177/1747493016681020
Shi Z, Miao C, Schoepf UJ et al (2020) A clinically applicable deep-learning model for detecting intracranial aneurysm in computed tomography angiography images. Nat Commun. 11:6090. https://doi.org/10.1038/s41467-020-19527-w
doi: 10.1038/s41467-020-19527-w
pubmed: 33257700
pmcid: 7705757
Winkel DJ, Heye T, Weikert TJ et al (2019) Evaluation of an AI-based detection software for acute findings in abdominal computed tomography scans: toward an automated work list prioritization of routine CT examinations. Invest Radiol 54:55–59
doi: 10.1097/RLI.0000000000000509
Gorincour G, Monneuse O, Ben CA et al (2021) Management of abdominal emergencies in adults using telemedicine and artificial intelligence. J Visc Surg 158:S26–S31. https://doi.org/10.1016/j.jviscsurg.2021.01.008
doi: 10.1016/j.jviscsurg.2021.01.008
pubmed: 33714710
Weikert T, Winkel DJ, Bremerich J et al (2020) Automated detection of pulmonary embolism in CT pulmonary angiograms using an AI-powered algorithm. Eur Radiol. 30(12):6545–6553. https://doi.org/10.1007/s00330-020-06998-0
doi: 10.1007/s00330-020-06998-0
pubmed: 32621243
Prokop M, van Everdingen W, van Rees VT et al (2020) CO-RADS: a categorical CT assessment scheme for patients suspected of having COVID-19-definition and evaluation. Radiology 296(2):E97–E104
doi: 10.1148/radiol.2020201473
Société Française de Radiologie (2018) Qualité et sécurité des actes de téléimagerie – Guide de bonnes pratiques. http://www.sfrnet.org/sfr/professionnels/2-infos-professionnelles/05-teleradiologie/index.phtml . Accessed 4 Apr 2020
AIDOC (2020) Pulmonary embolism guidelines and the intersection with AI. https://www.aidoc.com/blog/pulmonary-embolism-guidelines-and-the-intersection-with-ai/ . Accessed 1 June 2020
Maizlin ZV, Vos PM, Godoy MB, Cooperberg PL (2007) Computer-aided Detection of Pulmonary Embolism on CT Angiography Initial Experience. J Thorac Imaging 22:324–329. https://doi.org/10.1097/RTI.0b013e31815b89ca
doi: 10.1097/RTI.0b013e31815b89ca
pubmed: 18043386
Bouma H, Sonnemans JJ, Vilanova A, Gerritsen FA (2009) Automatic detection of pulmonary embolism in CTA images. IEEE Trans Med Imaging 28:1223–1230. https://doi.org/10.1109/TMI.2009.2013618
doi: 10.1109/TMI.2009.2013618
pubmed: 19211341
Brown JB, Gestring ML, Leeper CM et al (2017) The value of the injury severity score in pediatric trauma: time for a new definition of severe injury? J Trauma Acute Care Surg. https://doi.org/10.1097/TA.0000000000001440
Lee CW, Seo JB, Song J et al (2011) Evaluation of computer-aided detection and dual energy software in detection of peripheral pulmonary embolism on dual-energy pulmonary CT angiography. Eur Radiol:54–62
Lee G, Lee HY, Park H et al (2016) Radiomics and its emerging role in lung cancer research, imaging biomarkers and clinical management : state of the art. Eur J Radiol. https://doi.org/10.1016/j.ejrad.2016.09.005
Masutani Y, Macmahon H, Doi K (2002) Computerized detection of pulmonary embolism in spiral CT angiography based on volumetric image analysis. IEEE Trans Med Imaging 21:1517–1523. https://doi.org/10.1109/TMI.2002.806586
doi: 10.1109/TMI.2002.806586
pubmed: 12588035
Pichon E, Novak CL, Kiraly AP, Naidich DP (2004) A novel method for pulmonary emboli visualization from high-resolution CT images. Proc. SPIE 5367, Medical Imaging 2004: Visualization, Image-Guided Procedures, and Display, (5 May 2004); Robert L. Galloway Jr., Editor(s). https://doi.org/10.1117/12.532892
Liang J, Bi J (2007) Computer aided detection of pulmonary embolism with tobogganing and multiple instance classification in CT pulmonary angiography. Inf Process Med Imaging 20:630–641
pubmed: 17633735
Digumarthy SR, Kagay CR, Legasto AC, et al (2006) Computer-aided detection (CAD) of acute pulmonary emboli: evaluation in patients without significant pulmonary disease. In: Radiological Society of North America 2006 Scientific Assembly and Annual Meeting. Chicago IL
Schoepf UJ, Schneider AC, Das M et al (2007) Pulmonary embolism : computer-aided detection at multidetector row spiral computed tomography. J Thorac Imaging 22:319–323
doi: 10.1097/RTI.0b013e31815842a9
Wittenberg R, Peters JF, Sonnemans JJ et al (2010) Computer-assisted detection of pulmonary embolism : evaluation of pulmonary CT angiograms performed in an on-call setting. Eur Radiol 20:801–806
doi: 10.1007/s00330-009-1628-7
Blackmon KN, Florin C, Bogoni L et al (2011) Computer-aided detection of pulmonary embolism at CT pulmonary angiography : can it improve performance of inexperienced readers ? Eur Radiol 21:1214–1223
doi: 10.1007/s00330-010-2050-x
Wittenberg R, Berger FH, Peters JF et al (2012) Acute pulmonary embolism : effect of a computer-assisted detection prototype on purpose: methods: results. Radiology 262. https://doi.org/10.1148/radiol.11110372
Lahiji K, Kligerman S, Jeudy J, White C (2014) Improved Accuracy of pulmonary embolism computer-aided detection using iterative reconstruction compared with filtered back projection. AJR Am J Roentgenol:763–771
Wittenberg R, Peters JF, Van Den BIAH et al (2013) Computed tomography pulmonary angiography in acute pulmonary embolism -the effect of a computer-assisted detection prototype used as a concurrent reader. J Thorac Imaging 28:315–321
doi: 10.1097/RTI.0b013e3182870b97
Bhargavan M, Kaye AH, Forman HP, Sunshine JH (2009) Workload of radiologists in United States in 2006–2007 and trends since 1991–1992. Radiology 252. https://doi.org/10.1148/radiol.2522081895
Calder KK, Herbert M, Henderson SO (2005) The mortality of untreated pulmonary embolism in emergency department patients. Ann Emerg Med 45:302–310
doi: 10.1016/j.annemergmed.2004.10.001
Das M, Mühlenbruch G, Helm A et al (2008) Computer-aided detection of pulmonary embolism : influence on radiologists ’ detection performance with respect to vessel segments. Eur Radiol:1350–1355
Miller WTJ, Arinari LA, Barbosa EJ et al (2015) Small pulmonary artery defects are not reliable indicators of pulmonary embolism. Ann Am Thorac Soc. https://doi.org/10.1513/AnnalsATS.201502-105OC
Leithner D, Gruber-Rouh T, Beeres M et al (2018) 90-kVp low-tube-voltage CT pulmonary angiography in combination with advanced modeled iterative reconstruction algorithm: Effects on radiation dose, image quality and diagnostic accuracy for the detection of pulmonary embolism. Br J Radiol 91. https://doi.org/10.1259/bjr.20180269
Lysdahlgaard S, Hess S, Gerke O, Weber Kusk M (2020) A systematic literature review and meta-analysis of spectral CT compared to scintigraphy in the diagnosis of acute and chronic pulmonary embolisms. Eur Radiol 30:3624–3633. https://doi.org/10.1007/s00330-020-06735-7
doi: 10.1007/s00330-020-06735-7
pubmed: 32112117
Crombé A, Lecomte J-C, Banaste N et al (2021) Emergency teleradiological activity is an epidemiological estimator and predictor of the COVID-10 pandemic in mainland France. Insights Imaging 12(1):103
doi: 10.1186/s13244-021-01040-3
Grillet F, Busse-Coté A, Calame P et al (2020) COVID-19 pneumonia: microvascular disease revealed on pulmonary dual-energy computed tomography angiography. Quant Imaging Med Surg 10:1852–1862
doi: 10.21037/qims-20-708