Paper Info
Title
Preliminary study of generalized semiautomatic segmentation for 3D voxel labeling of lesions based on deep learning
Abstract
Purpose The three-dimensional (3D) voxel labeling of lesions requires significant radiologists' effort in the development of computer-aided detection software. To reduce the time required for the 3D voxel labeling, we aimed to develop a generalized semiautomatic segmentation method based on deep learning via a data augmentation-based domain generalization framework. In this study, we investigated whether a generalized semiautomatic segmentation model trained using two types of lesion can segment previously unseen types of lesion. Methods We targeted lung nodules in chest CT images, liver lesions in hepatobiliary-phase images of Gd-EOB-DTPA-enhanced MR imaging, and brain metastases in contrast-enhanced MR images. For each lesion, the 32 x 32 x 32 isotropic volume of interest (VOI) around the center of gravity of the lesion was extracted. The VOI was input into a 3D U-Net model to define the label of the lesion. For each type of target lesion, we compared five types of data augmentation and two types of input data. Results For all considered target lesions, the highest dice coefficients among the training patterns were obtained when using a combination of the existing data augmentation-based domain generalization framework and random monochrome inversion and when using the resized VOI as the input image. The dice coefficients were 0.639 +/- 0.124 for the lung nodules, 0.660 +/- 0.137 for the liver lesions, and 0.727 +/- 0.115 for the brain metastases. Conclusions Our generalized semiautomatic segmentation model could label unseen three types of lesion with different contrasts from the surroundings. In addition, the resized VOI as the input image enables the adaptation to the various sizes of lesions even when the size distribution differed between the training set and the test set.
Year
DOI
Venue
2021
10.1007/s11548-021-02504-z
INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY
Keywords
DocType
Volume
Semiautomatic segmentation, Deep learning, Domain generalization, Data augmentation, Computer-aided detection (CAD)
Journal
16
Issue
ISSN
Citations 
11
1861-6410
0
PageRank 
References 
Authors
0.34
0
10
Name
Order
Citations
PageRank
Nomura, Y.1319.51
Shouhei Hanaoka2267.56
Tomomi Takenaga342.58
Takahiro Nakao432.49
Hisaichi Shibata501.01
Soichiro Miki6156.44
Takeharu Yoshikawa7267.93
Takeyuki Watadani810.70
Naoto Hayashi9206.38
Osamu Abe1096.36