Screening of adolescent idiopathic scoliosis using generative adversarial network (GAN) inversion method in chest radiographs
PLOS ONE · 2023
DOI: 10.1371/journal.pone.0285489 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, 2023. Verfasst von Jun Soo Lee und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Industrial Engineering, Seoul National University, Seoul, Korea. Hauptthemen laut MEDLINE: Scoliosis, Kyphosis. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Education.
Abstract (Originalsprache)
Objective
Conventional computer-aided diagnosis using convolutional neural networks (CNN) has limitations in detecting sensitive changes and determining accurate decision boundaries in spectral and structural diseases such as scoliosis. We devised a new method to detect and diagnose adolescent idiopathic scoliosis in chest X-rays (CXRs) employing the latent space’s discriminative ability in the generative adversarial network (GAN) and a simple multi-layer perceptron (MLP) to screen adolescent idiopathic scoliosis CXRs. Materials and methods Our model was trained and validated in a two-step manner. First, we trained a GAN using CXRs with various scoliosis severities and utilized the trained network as a feature extractor using the GAN inversion method. Second, we classified each vector from the latent space using a simple MLP.
Results
The 2-layer MLP exhibited the best classification in the ablation study. With this model, the area under the receiver operating characteristic (AUROC) curves were 0.850 in the internal and 0.847 in the external datasets. Furthermore, when the sensitivity was fixed at 0.9, the model’s specificity was 0.697 in the internal and 0.646 in the external datasets.
Conclusion
We developed a classifier for Adolescent idiopathic scoliosis (AIS) through generative representation learning. Our model shows good AUROC under screening chest radiographs in both the internal and external datasets. Our model has learned the spectral severity of AIS, enabling it to generate normal images even when trained solely on scoliosis radiographs.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Industrial Engineering, Seoul National University, Seoul, Korea.
- Department of Biomedical Engineering, Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea.
- Department of Orthopedic Surgery, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Korea.
- Department of Radiology, Hanyang University Hospital, Seoul, Korea.
- Department of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea.
- Department of Convergence Medicine, Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea.
ORCID-Profile: Jun Soo Lee, Keewon Shin, Seung Min Ryu, Seong Gyu Jegal, Woojin Lee, Namkug Kim
Förderung
- Ministry of Education
Zitieren
APA
Lee, J. S., Shin, K., Ryu, S. M., Jegal, S. G., Lee, W., Yoon, M. A., Hong, G. S., Paik, S., & Kim, N. (2023). Screening of adolescent idiopathic scoliosis using generative adversarial network (GAN) inversion method in chest radiographs. PLOS ONE. https://doi.org/10.1371/journal.pone.0285489
Vancouver
Lee JS, Shin K, Ryu SM, Jegal SG, Lee W, Yoon MA, et al. Screening of adolescent idiopathic scoliosis using generative adversarial network (GAN) inversion method in chest radiographs. PLOS ONE. 2023. doi:10.1371/journal.pone.0285489
BibTeX
@article{Lee2023,
author = {Lee, Jun Soo and Shin, Keewon and Ryu, Seung Min and Jegal, Seong Gyu and Lee, Woojin and Yoon, Min A and Hong, Gil-Sun and Paik, Sanghyun and Kim, Namkug},
title = {Screening of adolescent idiopathic scoliosis using generative adversarial network (GAN) inversion method in chest radiographs},
journal = {PLOS ONE},
year = {2023},
doi = {10.1371/journal.pone.0285489}
}
EUVIMED