-
Tytuł:
-
Diagnostic performance for detecting bone marrow edema of the hip on dual-energy CT: Deep learning model vs. musculoskeletal physicians and radiologists.
-
Autorzy:
-
Park C; School of Biomedical Convergence Engineering, College of Information and BioMedical Engineering, Pusan National University, Yangsan, Korea.
Kim M; School of Biomedical Convergence Engineering, College of Information and BioMedical Engineering, Pusan National University, Yangsan, Korea.
Park C; Department of Radiology, Research Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Korea. Electronic address: .
Son W; Department of Radiology, Research Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Korea.
Lee SM; Department of Orthopedics, Research Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Korea.
Seok Jeong H; Department of Radiology, Research Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Korea.
Kang J; School of Biomedical Convergence Engineering, College of Information and BioMedical Engineering, Pusan National University, Yangsan, Korea.
Choi MH; Department of Preventive and Occupational & Environmental Medicine, Pusan National University Yangsan Hospital, Pusan National University, Yangsan, Korea.
-
Źródło:
-
European journal of radiology [Eur J Radiol] 2022 Jul; Vol. 152, pp. 110337. Date of Electronic Publication: 2022 Apr 30.
-
Typ publikacji:
-
Journal Article
-
Język:
-
English
-
Imprint Name(s):
-
Publication: Limerick : Elsevier Science Ireland Ltd
Original Publication: Stuttgart ; New York : Thieme, [c1981-
-
MeSH Terms:
-
Bone Marrow Diseases*/diagnostic imaging
Deep Learning*
Adult ; Aged ; Bone Marrow/diagnostic imaging ; Edema/diagnostic imaging ; Female ; Humans ; Middle Aged ; Radiologists ; Retrospective Studies ; Sensitivity and Specificity ; Tomography, X-Ray Computed/methods
-
Contributed Indexing:
-
Keywords: Bone marrow edema; Convolutional neural network; Deep learning; Diagnostic performance; Dual-energy CT
-
Entry Date(s):
-
Date Created: 20220507 Date Completed: 20220608 Latest Revision: 20220608
-
Update Code:
-
20240105
-
DOI:
-
10.1016/j.ejrad.2022.110337
-
PMID:
-
35525130
-
Purpose: To compare the diagnostic performance of a deep learning (DL) model with that of musculoskeletal physicians and radiologists for detecting bone marrow edema on dual-energy CT (DECT).
Method: This retrospective study included adult patients underwent hip DECT and MRI within 1 month between April 2018 and December 2020. A total of 8709 DECT images were divided into training/validation (85%, 7412 augmented images) and test (15%, 1297 images) sets. The images were labeled as present/absent bone marrow edema, with MRI as reference standard. We developed and trained a DL model to detect bone marrow edema from DECT images. Thereafter, DL model, two orthopedic surgeons, and three radiologists evaluated the presence of bone marrow edema on every test image. The diagnostic performance of the DL model and readers was compared. Inter-reader agreement was calculated using Fleiss-kappa statistics.
Results: A total of 73 patients (mean age, 59 ± 12 years; 38 female) were included. The DL model had a significantly higher area under the curve (AUC, 0.84 vs. 0.61-0.70, p < 0.001) and sensitivity (79% vs. 29-66%) without loss of specificity (90% vs. 74-93%) than the non- or less-experienced readers and similar to the trained reader (AUC, 0.83, p = 0.402; sensitivity, 71%; specificity, 94%). Additionally, AUCs were strongly dependent on the reader's DECT experience. Inter-reader agreement was fair (κ = 0.303).
Conclusion: The DL model showed better diagnostic performance than less-experienced physicians in detecting bone marrow edema on DECT and comparable performance to a trained radiologist.
(Copyright © 2022 Elsevier B.V. All rights reserved.)
