Deep learning-based decision forest for hereditary clear cell renal cell carcinoma segmentation on MRI.
MRI
decision forest
deep learning
segmentation
von Hippel-Lindau
Journal
Medical physics
ISSN: 2473-4209
Titre abrégé: Med Phys
Pays: United States
ID NLM: 0425746
Informations de publication
Date de publication:
Aug 2023
Aug 2023
Historique:
revised:
15
12
2022
received:
18
07
2022
accepted:
04
01
2023
medline:
15
8
2023
pubmed:
2
3
2023
entrez:
1
3
2023
Statut:
ppublish
Résumé
von Hippel-Lindau syndrome (VHL) is an autosomal dominant hereditary syndrome with an increased predisposition of developing numerous cysts and tumors, almost exclusively clear cell renal cell carcinoma (ccRCC). Considering the lifelong surveillance in such patients to monitor the disease, patients with VHL are preferentially imaged using MRI to eliminate radiation exposure. Segmentation of kidney and tumor structures on MRI in VHL patients is useful in lesion characterization (e.g., cyst vs. tumor), volumetric lesion analysis, and tumor growth prediction. However, automated tasks such as ccRCC segmentation on MRI is sparsely studied. We develop segmentation methodology for ccRCC on T1 weighted precontrast, corticomedullary, nephrogenic, and excretory contrast phase MRI. We applied a new neural network approache using a novel differentiable decision forest, called hinge forest (HF), to segment kidney parenchyma, cyst, and ccRCC tumors in 117 images from 115 patients. This data set represented an unprecedented 504 ccRCCs with 1171 cystic lesions obtained at five different MRI scanners. The HF architecture was compared with U-Net on 10 randomized splits with 75% used for training and 25% used for testing. Both methods were trained with Adam using default parameters ( The HF achieved an average kidney, cyst, and tumor Dice similarity coefficient (DSC) of 0.75 ± 0.03, 0.44 ± 0.05, 0.53 ± 0.04, respectively, while U-Net achieved an average kidney, cyst, and tumor DSC of 0.78 ± 0.02, 0.41 ± 0.04, 0.46 ± 0.05, respectively. The HF significantly outperformed U-Net on tumors while U-Net significantly outperformed HF when segmenting kidney parenchymas ( For the task of ccRCC segmentation, the HF can offer better segmentation performance compared to the traditional U-Net architecture. The leaf maps can glean hints about deep learning features that might prove to be useful in other automated tasks such as tumor characterization.
Sections du résumé
BACKGROUND
BACKGROUND
von Hippel-Lindau syndrome (VHL) is an autosomal dominant hereditary syndrome with an increased predisposition of developing numerous cysts and tumors, almost exclusively clear cell renal cell carcinoma (ccRCC). Considering the lifelong surveillance in such patients to monitor the disease, patients with VHL are preferentially imaged using MRI to eliminate radiation exposure.
PURPOSE
OBJECTIVE
Segmentation of kidney and tumor structures on MRI in VHL patients is useful in lesion characterization (e.g., cyst vs. tumor), volumetric lesion analysis, and tumor growth prediction. However, automated tasks such as ccRCC segmentation on MRI is sparsely studied. We develop segmentation methodology for ccRCC on T1 weighted precontrast, corticomedullary, nephrogenic, and excretory contrast phase MRI.
METHODS
METHODS
We applied a new neural network approache using a novel differentiable decision forest, called hinge forest (HF), to segment kidney parenchyma, cyst, and ccRCC tumors in 117 images from 115 patients. This data set represented an unprecedented 504 ccRCCs with 1171 cystic lesions obtained at five different MRI scanners. The HF architecture was compared with U-Net on 10 randomized splits with 75% used for training and 25% used for testing. Both methods were trained with Adam using default parameters (
RESULTS
RESULTS
The HF achieved an average kidney, cyst, and tumor Dice similarity coefficient (DSC) of 0.75 ± 0.03, 0.44 ± 0.05, 0.53 ± 0.04, respectively, while U-Net achieved an average kidney, cyst, and tumor DSC of 0.78 ± 0.02, 0.41 ± 0.04, 0.46 ± 0.05, respectively. The HF significantly outperformed U-Net on tumors while U-Net significantly outperformed HF when segmenting kidney parenchymas (
CONCLUSIONS
CONCLUSIONS
For the task of ccRCC segmentation, the HF can offer better segmentation performance compared to the traditional U-Net architecture. The leaf maps can glean hints about deep learning features that might prove to be useful in other automated tasks such as tumor characterization.
Identifiants
pubmed: 36855860
doi: 10.1002/mp.16303
pmc: PMC10683486
mid: NIHMS1943784
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
5020-5029Subventions
Organisme : Intramural NIH HHS
ID : ZIA BC011028
Pays : United States
Organisme : Intramural NIH HHS
ID : ZIH BC012032
Pays : United States
Informations de copyright
Published 2023. This article is a U.S. Government work and is in the public domain in the USA.
Références
Eur J Hum Genet. 2011 Jun;19(6):617-23
pubmed: 21386872
IEEE Trans Med Imaging. 2009 Aug;28(8):1251-65
pubmed: 19211338
Med Image Anal. 2021 Jan;67:101821
pubmed: 33049579
Med Image Anal. 2019 Apr;53:197-207
pubmed: 30802813
Magn Reson Med. 1995 Dec;34(6):910-4
pubmed: 8598820
Comput Biol Med. 2019 Jun;109:218-225
pubmed: 31078126