Deep Learning-based Diagnosis and Localization of Pneumothorax on Portable Supine Chest X-ray in Intensive and Emergency Medicine: A Retrospective Study.
Purpose: To develop two deep learning-based systems for diagnosing and localizing pneumothorax on portable supine chest X-rays (SCXRs). Methods: For this retrospective study, images meeting the following inclusion criteria were included: (1) patient age ≥ 20 years; (2) portable SCXR; (3) imaging obt...
| Publicado en: | Journal of Medical Systems Vol. 48; no. 1; pp. 1 - 11 |
|---|---|
| Autores principales: | , , , , , , , , , , , |
| Formato: | algorithm diagnostic images research tables/charts Journal Article |
| Publicado: |
Springer Nature
12/4/2023
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=174877250&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 174877250 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 01485598 4N0 jtl: Journal of Medical Systems issn: 01485598 maglogo: N pubinfo: dt: 12/4/2023 vid: 48 iid: 1 pid: 237 pub: Springer Nature place: New York, New York artinfo: ui: 174877250 174877250 174877250 10.1007/s10916-023-02023-1 174877250 ppf: 1 ppct: 10 formats: fmt: – @attributes: type: T – @attributes: type: P tig: atl: Deep Learning-based Diagnosis and Localization of Pneumothorax on Portable Supine Chest X-ray in Intensive and Emergency Medicine: A Retrospective Study. aug: au: Wang, Chih-Hung Lin, Tzuching Chen, Guanru Lee, Meng-Rui Tay, Joyce Wu, Cheng-Yi Wu, Meng-Che Roth, Holger R. Yang, Dong Zhao, Can Wang, Weichung Huang, Chien-Hua affil: https://ror.org/05bqach95 Department of Emergency Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan sug: subj: Deep Learning Utilization Pneumothorax Diagnosis Supine Position Radiography, Thoracic Emergency Service Intensive Care Units Human Retrospective Design ROC Curve Pneumothorax Risk Factors Prediction Models Middle Age Aged Male Female Paired T-Tests Descriptive Statistics Confidence Intervals Two-Tailed Test Middle Aged: 45-64 years Aged: 65+ years Male Female ab: Purpose: To develop two deep learning-based systems for diagnosing and localizing pneumothorax on portable supine chest X-rays (SCXRs). Methods: For this retrospective study, images meeting the following inclusion criteria were included: (1) patient age ≥ 20 years; (2) portable SCXR; (3) imaging obtained in the emergency department or intensive care unit. Included images were temporally split into training (1571 images, between January 2015 and December 2019) and testing (1071 images, between January 2020 to December 2020) datasets. All images were annotated using pixel-level labels. Object detection and image segmentation were adopted to develop separate systems. For the detection-based system, EfficientNet-B2, DneseNet-121, and Inception-v3 were the architecture for the classification model; Deformable DETR, TOOD, and VFNet were the architecture for the localization model. Both classification and localization models of the segmentation-based system shared the UNet architecture. Results: In diagnosing pneumothorax, performance was excellent for both detection-based (Area under receiver operating characteristics curve [AUC]: 0.940, 95% confidence interval [CI]: 0.907–0.967) and segmentation-based (AUC: 0.979, 95% CI: 0.963–0.991) systems. For images with both predicted and ground-truth pneumothorax, lesion localization was highly accurate (detection-based Dice coefficient: 0.758, 95% CI: 0.707–0.806; segmentation-based Dice coefficient: 0.681, 95% CI: 0.642–0.721). The performance of the two deep learning-based systems declined as pneumothorax size diminished. Nonetheless, both systems were similar or better than human readers in diagnosis or localization performance across all sizes of pneumothorax. Conclusions: Both deep learning-based systems excelled when tested in a temporally different dataset with differing patient or image characteristics, showing favourable potential for external generalizability. pubtype: Academic Journal doctype: algorithm diagnostic images research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|