Detection and severity quantification of pulmonary embolism with 3D CT data using an automated deep learning-based artificial solution

被引：6

作者：

Djahnine, Aissam ^{[1
,2
]}

Lazarus, Carole ^{[1
]}

Lederlin, Mathieu ^{[3
]}

Mule, Sebastien ^{[4
,5
]}

Wiemker, Rafael ^{[1
]}

Si-Mohamed, Salim ^{[6
]}

Jupin-Delevaux, Emilien ^{[6
]}

Nempont, Olivier ^{[1
]}

Skandarani, Youssef ^{[1
]}

De Craene, Mathieu ^{[1
]}

Goubalan, Segbedji ^{[1
]}

Raynaud, Caroline ^{[1
]}

Belkouchi, Younes ^{[7
,8
]}

Ben Afia, Amira ^{[8
]}

Fabre, Clement ^{[10
]}

Ferretti, Gilbert ^{[11
]}

De Margerie, Constance ^{[12
,13
]}

Berge, Pierre ^{[14
]}

Liberge, Renan ^{[15
]}

Elbaz, Nicolas ^{[16
]}

Blain, Maxime ^{[17
]}

Brillet, Pierre -Yves ^{[18
]}

Chassagnon, Guillaume ^{[12
,19
]}

Cadour, Farah ^{[20
]}

Caramella, Caroline ^{[21
]}

El Hajjam, Mostafa ^{[22
]}

Boussouar, Samia ^{[23
]}

Hadchiti, Joya ^{[24
]}

Fablet, Xavier ^{[3
]}

Khalil, Antoine ^{[9
]}

Talbot, Hugues ^{[8
]}

Luciani, Alain ^{[4
,5
]}

Lassau, Nathalie ^{[24
]}

Boussel, Loic ^{[2
,6
]}

机构：

[1] Philips Res France, F-92150 Suresnes, France

[2] Univ Claude Bernard Lyon 1, CREATIS, INSA Lyon, UJM St Etienne,NRS,Inserm,CREATIS UMR 5220,U1294, Lyon, France

[3] CHU Rennes, Dept Radiol, F-35000 Rennes, France

[4] Henri Mondor Univ Hosp, Med Imaging Dept, AP HP, Creteil, France

[5] Inserm, Team 18, U955, F-94000 Creteil, France

[6] Hosp Civils Lyon, Dept Radiol, F-69500 Lyon, France

[7] Univ Paris Saclay, BIOMAPS, CNRS,CEA,UMR 1281, Inserm,Lab Imagerie Biomed Multimodale Paris Sacl, F-94800 Villejuif, France

[8] Univ Paris Saclay, CVN Ctr vis Numer, OPTS Optimisat Imagerie & Sante, Inria,CentraleSupelec, F-91190 Gif Sur Yvette, France

[9] Hop Bichat Claude Bernard, Dept Radiol, APHP Nord, F-75018 Paris, France

[10] Ctr Hosp Laval, Dept Radiol, F-53000 Laval, France

[11] Univ Grenobles Alpes, Serv Radiol & Imagerie Med, F-38000 Grenoble, France

[12] Univ Paris Cite, F-75006 Paris, France

[13] Hop St Louis, AP HP, Dept Radiol, F-75010 Paris, France

[14] CHU Angers, Dept Radiol, Angers, France

[15] CHU Nantes, Dept Radiol, F-44000 Nantes, France

[16] Hop Europeen Georges Pompidou, Dept Radiol, AP HP, F-75015 Paris, France

[17] Hop Henri Mondor, AP HP, Dept Radiol, F-94000 Creteil, France

[18] Paris 13 Univ, Hop Avicenne, Dept Radiol, F-93000 Bobigny, France

[19] Hop Cochin, Dept Radiol, APHP, F-75014 Paris, France

[20] Hop Univ Timone, APHM, CEMEREM, F-13005 Marseille, France

[21] Grp Hosp Paris St Joseph, Dept Radiol, F-75015 Paris, France

[22] Hop Ambroise Pare Hosp, Dept Radiol, Team 3, UVSQ,UMR 1179,INSERM, Boulogne Billancourt, France

[23] Sorbonne Univ, Hop La Pitie Salpetriere, Unite Imagerie Cardiovasc & Thorac ICT, UMRS 1166, F-75013 Paris, France

[24] Univ Paris Saclay, Inst Gustave Roussy, Dept Imaging, F-94800 Villejuif, France

来源：

DIAGNOSTIC AND INTERVENTIONAL IMAGING | 2024年 / 105卷 / 03期

关键词：

Artificial intelligence; Pulmonary embolism; Qanadli score; Retina U-net; R-CNN; INTELLIGENCE;

D O I：

10.1016/j.diii.2023.09.006

中图分类号：

R8 [特种医学]; R445 [影像诊断学];

学科分类号：

1002 ; 100207 ; 1009 ;

摘要：

Purpose: The purpose of this study was to propose a deep learning-based approach to detect pulmonary embolism and quantify its severity using the Qanadli score and the right-to-left ventricle diameter (RV/LV) ratio on three-dimensional (3D) computed tomography pulmonary angiography (CTPA) examinations with limited annotations. Materials and methods: Using a database of 3D CTPA examinations of 1268 patients with image-level annotations, and two other public datasets of CTPA examinations from 91 (CAD-PE) and 35 (FUME-PE) patients with pixel-level annotations, a pipeline consisting of: (i), detecting blood clots; (ii), performing PE-positive versus negative classification; (iii), estimating the Qanadli score; and (iv), predicting RV/LV diameter ratio was followed. The method was evaluated on a test set including 378 patients. The performance of PE classification and severity quantification was quantitatively assessed using an area under the curve (AUC) analysis for PE classification and a coefficient of determination (R2) for the Qanadli score and the RV/LV diameter ratio. Results: Quantitative evaluation led to an overall AUC of 0.870 (95% confidence interval [CI]: 0.850-0.900) for PE classification task on the training set and an AUC of 0.852 (95% CI: 0.810-0.890) on the test set. Regression analysis yielded R2 value of 0.717 (95% CI: 0.668-0.760) and of 0.723 (95% CI: 0.668-0.766) for the Qanadli score and the RV/LV diameter ratio estimation, respectively on the test set. Conclusion: This study shows the feasibility of utilizing AI-based assistance tools in detecting blood clots and estimating PE severity scores with 3D CTPA examinations. This is achieved by leveraging blood clots and cardiac segmentations. Further studies are needed to assess the effectiveness of these tools in clinical practice. (c) 2023 Societe francaise de radiologie. Published by Elsevier Masson SAS. All rights reserved.

引用

页码：97 / 103

页数：7

共 50 条

[1] Automated detection of pulmonary embolism from CT-angiograms using deep learning
Heidi Huhtanen
Mikko Nyman
Tarek Mohsen
Arho Virkki
Antti Karlsson
Jussi Hirvonen
BMC Medical Imaging, 22
[2] Automated detection of pulmonary embolism from CT-angiograms using deep learning
Huhtanen, Heidi
Nyman, Mikko
Mohsen, Tarek
Virkki, Arho
Karlsson, Antti
Hirvonen, Jussi
BMC MEDICAL IMAGING, 2022, 22 (01)
[3] Deep Learning-Based Algorithm for Automatic Detection of Pulmonary Embolism in Chest CT Angiograms
Grenier, Philippe A.
Ayobi, Angela
Quenet, Sarah
Tassy, Maxime
Marx, Michael
Chow, Daniel S.
Weinberg, Brent D.
Chang, Peter D.
Chaibi, Yasmina
DIAGNOSTICS, 2023, 13 (07)
[4] Generalizability assessment of COVID-19 3D CT data for deep learning-based disease detection
Fallahpoor, Maryam
Chakraborty, Subrata
Heshejin, Mohammad Tavakoli
Chegeni, Hossein
Horry, Michael James
Pradhan, Biswajeet
COMPUTERS IN BIOLOGY AND MEDICINE, 2022, 145
[5] nnU-Net-based deep-learning for pulmonary embolism: detection, clot volume quantification, and severity correlation in the RSPECT dataset
Lanza, Ezio
Ammirabile, Angela
Francone, Marco
EUROPEAN JOURNAL OF RADIOLOGY, 2024, 177
[6] Deep learning-based automated detection of pulmonary embolism on CT pulmonary angiograms: No significant effects on report communication times and patient turnaround in the emergency department nine months after technical implementation
Schmuelling, Lena
Franzeck, Fabian C.
Nickel, Christian H.
Mansella, Gregory
Bingisser, Roland
Schmidt, Noemi
Stieltjes, Bram
Bremerich, Jens
Sauter, Alexander W.
Weikert, Thomas
Sommer, Gregor
EUROPEAN JOURNAL OF RADIOLOGY, 2021, 141
[7] Detection and segmentation of pulmonary embolism in 3D CT pulmonary angiography using a threshold adjustment segmentation network
Fan, Jian-cong
Luan, Haoyang
Qiao, Yaqian
Li, Yang
Ren, Yande
SCIENTIFIC REPORTS, 2025, 15 (01):
[8] A deep learning-based algorithm improves radiology residents' diagnoses of acute pulmonary embolism on CT pulmonary angiograms
Vallee, Alexandre
Quint, Raphaelle
Brun, Anne Laure
Mellot, Francois
Grenier, Philippe A.
EUROPEAN JOURNAL OF RADIOLOGY, 2024, 171
[9] Automated measurement of penile curvature using deep learning-based novel quantification method
Baray, Sriman Bidhan
Abdelmoniem, Mohamed
Mahmud, Sakib
Kabir, Saidul
Faisal, Md. Ahasan Atick
Chowdhury, Muhammad E. H.
Abbas, Tariq O.
FRONTIERS IN PEDIATRICS, 2023, 11
[10] Deep learning for segmentation of 49 selected bones in CT scans: First step in automated PET/CT-based 3D quantification of skeletal metastases
Belal, Sarah Lindgren
Sadik, May
Kaboteh, Reza
Enqvist, Olof
Ulen, Johannes
Poulsen, Mads H.
Simonsen, Jane
Hoilund-Carlsen, Poul F.
Edenbrandt, Lars
Tragardh, Elin
EUROPEAN JOURNAL OF RADIOLOGY, 2019, 113 : 89 - 95

← 1 2 3 4 5 →