Automatically Detecting Pancreatic Cysts in Autosomal Dominant Polycystic Kidney Disease on MRI Using Deep Learning
Tomography · 2024
DOI: 10.3390/tomography10070087 ↗Kurz gefasst
Fachartikel in der Zeitschrift Tomography, 2024. Verfasst von Sophie J Wang und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Weill Cornell Medicine, New York, NY 10065, USA. Hauptthemen laut MEDLINE: Polycystic Kidney, Autosomal Dominant, Deep Learning, Pancreatic Cyst, Magnetic Resonance Imaging. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCATS NIH HHS, Weill Cornell Medical College (WCMC) Clinical and Translational Science Center (CTSC).
Abstract (Originalsprache)
Background
Pancreatic cysts in autosomal dominant polycystic kidney disease (ADPKD) correlate with PKD2 mutations, which have a different phenotype than PKD1 mutations. However, pancreatic cysts are commonly overlooked by radiologists. Here, we automate the detection of pancreatic cysts on abdominal MRI in ADPKD.
Methods
Eight nnU-Net-based segmentation models with 2D or 3D configuration and various loss functions were trained on positive-only or positive-and-negative datasets, comprising axial and coronal T2-weighted MR images from 254 scans on 146 ADPKD patients with pancreatic cysts labeled independently by two radiologists. Model performance was evaluated on test subjects unseen in training, comprising 40 internal, 40 external, and 23 test–retest reproducibility ADPKD patients.
Results
Two radiologists agreed on 52% of cysts labeled on training data, and 33%/25% on internal/external test datasets. The 2D model with a loss of combined dice similarity coefficient and cross-entropy trained with the dataset with both positive and negative cases produced an optimal dice score of 0.7 ± 0.5/0.8 ± 0.4 at the voxel level on internal/external validation and was thus used as the best-performing model. In the test–retest, the optimal model showed superior reproducibility (83% agreement between scan A and B) in segmenting pancreatic cysts compared to six expert observers (77% agreement). In the internal/external validation, the optimal model showed high specificity of 94%/100% but limited sensitivity of 20%/24%.
Conclusions
Labeling pancreatic cysts on T2 images of the abdomen in patients with ADPKD is challenging, deep learning can help the automated detection of pancreatic cysts, and further image quality improvement is warranted.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiology, Weill Cornell Medicine, New York, NY 10065, USA.
- The Rogosin Institute, New York, NY 10065, USA.
- Department of Medicine, Weill Cornell Medicine, New York, NY 10065, USA.
- Department of Radiology, Columbia University Vagelos College of Physicians and Surgeons, New York, NY 10032, USA.
ORCID-Profile: Sophie J Wang, Zhongxiu Hu, Collin Li, Xinzi He, Chenglin Zhu, Yin Wang, Usama Sattar, Vahid Bazojoo, Hui Yi Ng He, Jon D Blumenfeld, Martin R Prince
Förderung
- NCATS NIH HHS
- Weill Cornell Medical College (WCMC) Clinical and Translational Science Center (CTSC)
Zitieren
APA
Wang, S. J., Hu, Z., Li, C., He, X., Zhu, C., Wang, Y., Sattar, U., Bazojoo, V., He, H. Y. N., Blumenfeld, J. D., & Prince, M. R. (2024). Automatically Detecting Pancreatic Cysts in Autosomal Dominant Polycystic Kidney Disease on MRI Using Deep Learning. Tomography. https://doi.org/10.3390/tomography10070087
Vancouver
Wang SJ, Hu Z, Li C, He X, Zhu C, Wang Y, et al. Automatically Detecting Pancreatic Cysts in Autosomal Dominant Polycystic Kidney Disease on MRI Using Deep Learning. Tomography. 2024. doi:10.3390/tomography10070087
BibTeX
@article{Wang2024,
author = {Wang, Sophie J and Hu, Zhongxiu and Li, Collin and He, Xinzi and Zhu, Chenglin and Wang, Yin and Sattar, Usama and Bazojoo, Vahid and He, Hui Yi Ng and Blumenfeld, Jon D and Prince, Martin R},
title = {Automatically Detecting Pancreatic Cysts in Autosomal Dominant Polycystic Kidney Disease on MRI Using Deep Learning},
journal = {Tomography},
year = {2024},
doi = {10.3390/tomography10070087}
}
EUVIMED