Dual center validation of deep learning for automated multi-label segmentation of thoracic anatomy in bedside chest radiographs
Computer Methods and Programs in Biomedicine · 2023
DOI: 10.1016/j.cmpb.2023.107505 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computer Methods and Programs in Biomedicine, 2023. Verfasst von Felix Busch und 13 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt Universität zu Berlin, Berli…. Hauptthemen laut MEDLINE: Deep Learning. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 34 Literaturangaben. Gefördert von: Charité – Universitätsmedizin Berlin, Berlin Institute of Health.
Abstract (Originalsprache)
BACKGROUND AND OBJECTIVES: Bedside chest radiographs (CXRs) are challenging to interpret but important for monitoring cardiothoracic disease and invasive therapy devices in critical care and emergency medicine. Taking surrounding anatomy into account is likely to improve the diagnostic accuracy of artificial intelligence and bring its performance closer to that of a radiologist. Therefore, we aimed to develop a deep convolutional neural network for efficient automatic anatomy segmentation of bedside CXRs. METHODS: To improve the efficiency of the segmentation process, we introduced a "human-in-the-loop" segmentation workflow with an active learning approach, looking at five major anatomical structures in the chest (heart, lungs, mediastinum, trachea, and clavicles). This allowed us to decrease the time needed for segmentation by 32% and select the most complex cases to utilize human expert annotators efficiently. After annotation of 2,000 CXRs from different Level 1 medical centers at Charité - University Hospital Berlin, there was no relevant improvement in model performance, and the annotation process was stopped. A 5-layer U-ResNet was trained for 150 epochs using a combined soft Dice similarity coefficient (DSC) and cross-entropy as a loss function. DSC, Jaccard index (JI), Hausdorff distance (HD) in mm, and average symmetric surface distance (ASSD) in mm were used to assess model performance. External validation was performed using an independent external test dataset from Aachen University Hospital (n = 20). RESULTS: The final training, validation, and testing dataset consisted of 1900/50/50 segmentation masks for each anatomical structure. Our model achieved a mean DSC/JI/HD/ASSD of 0.93/0.88/32.1/5.8 for the lung, 0.92/0.86/21.65/4.85 for the mediastinum, 0.91/0.84/11.83/1.35 for the clavicles, 0.9/0.85/9.6/2.19 for the trachea, and 0.88/0.8/31.74/8.73 for the heart. Validation using the external dataset showed an overall robust performance of our algorithm. CONCLUSIONS: Using an efficient computer-aided segmentation method with active learning, our anatomy-based model achieves comparable performance to state-of-the-art approaches. Instead of only segmenting the non-overlapping portions of the organs, as previous studies did, a closer approximation to actual anatomy is achieved by segmenting along the natural anatomical borders. This novel anatomy approach could be useful for developing pathology models for accurate and quantifiable diagnosis.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiology, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt Universität zu Berlin, Berlin, Germany; Department of Anesthesiology, Division of Operative Intens…
- Department of Radiology, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt Universität zu Berlin, Berlin, Germany.
- Department of Diagnostic and Interventional Radiology, University Hospital Aachen, Aachen, Germany.
- Department of Radiology, Ludwig-Maximilians-University of Munich, Munich, Germany.
- Department of Radiology, Technical University of Munich, Munich, Germany.
- Berlin Institute of Health at Charité - Universitätsmedizin Berlin, Berlin, Germany; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA; Departments of Radia…
Förderung
- Charité – Universitätsmedizin Berlin
- Berlin Institute of Health
Zitieren
APA
Busch, F., Xu, L., Sushko, D., Weidlich, M., Truhn, D., Müller-Franzes, G., Heimer, M. M., Niehues, S. M., Makowski, M. R., Hinsche, M., Vahldiek, J. L., Aerts, H. J., Adams, L. C., & Bressem, K. K. (2023). Dual center validation of deep learning for automated multi-label segmentation of thoracic anatomy in bedside chest radiographs. Computer Methods and Programs in Biomedicine. https://doi.org/10.1016/j.cmpb.2023.107505
Vancouver
Busch F, Xu L, Sushko D, Weidlich M, Truhn D, Müller-Franzes G, et al. Dual center validation of deep learning for automated multi-label segmentation of thoracic anatomy in bedside chest radiographs. Computer Methods and Programs in Biomedicine. 2023. doi:10.1016/j.cmpb.2023.107505
BibTeX
@article{Busch2023,
author = {Busch, Felix and Xu, Lina and Sushko, Dmitry and Weidlich, Matthias and Truhn, Daniel and Müller-Franzes, Gustav and Heimer, Maurice M and Niehues, Stefan M and Makowski, Marcus R and Hinsche, Markus and Vahldiek, Janis L and Aerts, Hugo Jwl and Adams, Lisa C and Bressem, Keno K},
title = {Dual center validation of deep learning for automated multi-label segmentation of thoracic anatomy in bedside chest radiographs},
journal = {Computer Methods and Programs in Biomedicine},
year = {2023},
doi = {10.1016/j.cmpb.2023.107505}
}
EUVIMED