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

被引：5

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

[21] Survey on deep learning-based 3D object detection in autonomous driving
Liang, Zhenming
Huang, Yingping
TRANSACTIONS OF THE INSTITUTE OF MEASUREMENT AND CONTROL, 2023, 45 (04) : 761 - 776
[22] Improving 3D Metric GPR Imaging Using Automated Data Collection and Learning-Based Processing
Feng, Jinglun
Yang, Liang
Hoxha, Ejup
Xiao, Jizhong
IEEE SENSORS JOURNAL, 2023, 23 (05) : 4527 - 4539
[23] Automated Deep Learning-based Detection and Segmentation of Lung Tumors at CT Imaging
Kashyap, Mehr
Wang, Xi
Panjwani, Neil
Hasan, Mohammad
Zhang, Qin
Huang, Charles
Bush, Karl
Chin, Alexander
Vitzthum, Lucas K.
Dong, Peng
Zaky, Sandra
Loo, Billy W.
Diehn, Maximilian
Xing, Lei
Li, Ruijiang
Gensheimer, Michael F.
RADIOLOGY, 2025, 314 (01)
[24] Deep learning-based 3D reconstruction: a survey
Taha Samavati
Mohsen Soryani
Artificial Intelligence Review, 2023, 56 : 9175 - 9219
[25] Deep learning-based COVID-19 detection system using pulmonary CT scans
Nair, Rajit
Alhudhaif, Adi
Koundal, Deepika
Doewes, Rumi Iqbal
Sharma, Preeti
TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, 2021, 29 (29) : 2716 - 2727
[26] Deep learning-based 3D reconstruction: a survey
Samavati, Taha
Soryani, Mohsen
ARTIFICIAL INTELLIGENCE REVIEW, 2023, 56 (09) : 9175 - 9219
[27] AUTOMATED PULMONARY NODULE DETECTION USING 3D DEEP CONVOLUTIONAL NEURAL NETWORKS
Tang, Hao
Kim, Daniel R.
Xie, Xiaohui
2018 IEEE 15TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI 2018), 2018, : 523 - 526
[28] A learning-based crack defect detection and 3D localization framework for automated fluorescent
Wu, Qiang
Qin, Xunpen
Dong, Kang
Shi, Aixian
Hu, Zeqi
EXPERT SYSTEMS WITH APPLICATIONS, 2023, 214
[29] Learning-based 3D face detection using geometric context
Guo, Yanwen
Zhang, Fuyan
Liu, Chunxiao
Sun, Hanqiu
Peng, Qunsheng
COMPUTER ANIMATION AND VIRTUAL WORLDS, 2007, 18 (4-5) : 483 - 492
[30] Automated quantification of brain PET in PET/CT using deep learning-based CT-to-MR translation: a feasibility study
Kim, Daesung
Choo, Kyobin
Lee, Sangwon
Kang, Seongjin
Yun, Mijin
Yang, Jaewon
EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2025,

← 1 2 3 4 5 →