Patch-Based Deep-Learning Model With Limited Training Dataset for Liver Tumor Segmentation in Contrast-Enhanced Hepatic Computed Tomography

The automatic segmentation of liver tumors plays an important role in the diagnosis and treatment of liver cancer. While deep convolutional neural network (DCNN) models are widely used for segmentation tasks in medical imaging, they require 1,000 to 10,000 annotated cases for effective training. How...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuqiao Yang, Muneyuki Sato, Ze Jin, Kenji Suzuki
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11006069/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849761324673794048
author Yuqiao Yang
Muneyuki Sato
Ze Jin
Kenji Suzuki
author_facet Yuqiao Yang
Muneyuki Sato
Ze Jin
Kenji Suzuki
author_sort Yuqiao Yang
collection DOAJ
description The automatic segmentation of liver tumors plays an important role in the diagnosis and treatment of liver cancer. While deep convolutional neural network (DCNN) models are widely used for segmentation tasks in medical imaging, they require 1,000 to 10,000 annotated cases for effective training. However, assembling datasets of this size is a significant challenge due to the labor-intensive nature of tumor annotation, which often requires the expertise of radiologists. We propose a multi-scale Hessian-enhanced patch-based neural network, which we call an MHP-Net, for liver tumor segmentation with a limited dataset. Our approach involves sampling 3D patches from the input images for training a neural network, rather than using all input images, which are commonly used in DCNN training. We applied a multi-scale Hessian ellipsoid enhancer to extract multi-scale features of the liver tumor. We implemented a region-stratified sampling strategy to prevent overfitting in patch-based neural network training. We evaluated the effectiveness of our model using a dataset from the Liver Tumor Segmentation Benchmark (LiTS). To investigate the performance of the model under limited sample-size conditions, we trained it and state-of-the-art (SOTA) deep learning models with 7, 14, and 28 tumors. Our model achieved average Dice scores of 0.691, 0.709, and 0.719 which were higher than those ranging between 0.395 and 0.641 with the SOTA models. Remarkably, our model also achieved a Dice score (0.709) on par with the top model (0.702) in the MICCAI 2017 worldwide competition, despite utilizing only 1.5% (14 out of 908 tumors) of the training data.
format Article
id doaj-art-5c334e4ac5e7407596bfcd53a2b412fd
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-5c334e4ac5e7407596bfcd53a2b412fd2025-08-20T03:06:04ZengIEEEIEEE Access2169-35362025-01-0113868638687310.1109/ACCESS.2025.357072811006069Patch-Based Deep-Learning Model With Limited Training Dataset for Liver Tumor Segmentation in Contrast-Enhanced Hepatic Computed TomographyYuqiao Yang0https://orcid.org/0000-0003-0839-6989Muneyuki Sato1Ze Jin2Kenji Suzuki3https://orcid.org/0000-0002-3993-8309Biomedical Artificial Intelligence Research Unit (BMAI), Institute of Integrated Research, Institute of Science Tokyo, Tokyo, JapanBiomedical Artificial Intelligence Research Unit (BMAI), Institute of Integrated Research, Institute of Science Tokyo, Tokyo, JapanBiomedical Artificial Intelligence Research Unit (BMAI), Institute of Integrated Research, Institute of Science Tokyo, Tokyo, JapanBiomedical Artificial Intelligence Research Unit (BMAI), Institute of Integrated Research, Institute of Science Tokyo, Tokyo, JapanThe automatic segmentation of liver tumors plays an important role in the diagnosis and treatment of liver cancer. While deep convolutional neural network (DCNN) models are widely used for segmentation tasks in medical imaging, they require 1,000 to 10,000 annotated cases for effective training. However, assembling datasets of this size is a significant challenge due to the labor-intensive nature of tumor annotation, which often requires the expertise of radiologists. We propose a multi-scale Hessian-enhanced patch-based neural network, which we call an MHP-Net, for liver tumor segmentation with a limited dataset. Our approach involves sampling 3D patches from the input images for training a neural network, rather than using all input images, which are commonly used in DCNN training. We applied a multi-scale Hessian ellipsoid enhancer to extract multi-scale features of the liver tumor. We implemented a region-stratified sampling strategy to prevent overfitting in patch-based neural network training. We evaluated the effectiveness of our model using a dataset from the Liver Tumor Segmentation Benchmark (LiTS). To investigate the performance of the model under limited sample-size conditions, we trained it and state-of-the-art (SOTA) deep learning models with 7, 14, and 28 tumors. Our model achieved average Dice scores of 0.691, 0.709, and 0.719 which were higher than those ranging between 0.395 and 0.641 with the SOTA models. Remarkably, our model also achieved a Dice score (0.709) on par with the top model (0.702) in the MICCAI 2017 worldwide competition, despite utilizing only 1.5% (14 out of 908 tumors) of the training data.https://ieeexplore.ieee.org/document/11006069/Artificial intelligencedeep learningliver tumor segmentationmedical image analysissmall-data AI
spellingShingle Yuqiao Yang
Muneyuki Sato
Ze Jin
Kenji Suzuki
Patch-Based Deep-Learning Model With Limited Training Dataset for Liver Tumor Segmentation in Contrast-Enhanced Hepatic Computed Tomography
IEEE Access
Artificial intelligence
deep learning
liver tumor segmentation
medical image analysis
small-data AI
title Patch-Based Deep-Learning Model With Limited Training Dataset for Liver Tumor Segmentation in Contrast-Enhanced Hepatic Computed Tomography
title_full Patch-Based Deep-Learning Model With Limited Training Dataset for Liver Tumor Segmentation in Contrast-Enhanced Hepatic Computed Tomography
title_fullStr Patch-Based Deep-Learning Model With Limited Training Dataset for Liver Tumor Segmentation in Contrast-Enhanced Hepatic Computed Tomography
title_full_unstemmed Patch-Based Deep-Learning Model With Limited Training Dataset for Liver Tumor Segmentation in Contrast-Enhanced Hepatic Computed Tomography
title_short Patch-Based Deep-Learning Model With Limited Training Dataset for Liver Tumor Segmentation in Contrast-Enhanced Hepatic Computed Tomography
title_sort patch based deep learning model with limited training dataset for liver tumor segmentation in contrast enhanced hepatic computed tomography
topic Artificial intelligence
deep learning
liver tumor segmentation
medical image analysis
small-data AI
url https://ieeexplore.ieee.org/document/11006069/
work_keys_str_mv AT yuqiaoyang patchbaseddeeplearningmodelwithlimitedtrainingdatasetforlivertumorsegmentationincontrastenhancedhepaticcomputedtomography
AT muneyukisato patchbaseddeeplearningmodelwithlimitedtrainingdatasetforlivertumorsegmentationincontrastenhancedhepaticcomputedtomography
AT zejin patchbaseddeeplearningmodelwithlimitedtrainingdatasetforlivertumorsegmentationincontrastenhancedhepaticcomputedtomography
AT kenjisuzuki patchbaseddeeplearningmodelwithlimitedtrainingdatasetforlivertumorsegmentationincontrastenhancedhepaticcomputedtomography