Home > Research > Publications & Outputs > BDAL

Links

Text available via DOI:

View graph of relations

BDAL: Balanced Distribution Active Learning for MRI Cardiac Multistructures Segmentation

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

BDAL: Balanced Distribution Active Learning for MRI Cardiac Multistructures Segmentation. / Shu, Xiu; Yang, Yunyun; Liu, Jun et al.
In: IEEE Transactions on Industrial Informatics, Vol. 20, No. 4, 30.04.2024, p. 6099-6108.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Shu, X, Yang, Y, Liu, J, Chang, X & Wu, B 2024, 'BDAL: Balanced Distribution Active Learning for MRI Cardiac Multistructures Segmentation', IEEE Transactions on Industrial Informatics, vol. 20, no. 4, pp. 6099-6108. https://doi.org/10.1109/TII.2023.3342442

APA

Shu, X., Yang, Y., Liu, J., Chang, X., & Wu, B. (2024). BDAL: Balanced Distribution Active Learning for MRI Cardiac Multistructures Segmentation. IEEE Transactions on Industrial Informatics, 20(4), 6099-6108. https://doi.org/10.1109/TII.2023.3342442

Vancouver

Shu X, Yang Y, Liu J, Chang X, Wu B. BDAL: Balanced Distribution Active Learning for MRI Cardiac Multistructures Segmentation. IEEE Transactions on Industrial Informatics. 2024 Apr 30;20(4):6099-6108. Epub 2023 Dec 29. doi: 10.1109/TII.2023.3342442

Author

Shu, Xiu ; Yang, Yunyun ; Liu, Jun et al. / BDAL : Balanced Distribution Active Learning for MRI Cardiac Multistructures Segmentation. In: IEEE Transactions on Industrial Informatics. 2024 ; Vol. 20, No. 4. pp. 6099-6108.

Bibtex

@article{d6cacd7981ba448c919be15c80b0d14b,
title = "BDAL: Balanced Distribution Active Learning for MRI Cardiac Multistructures Segmentation",
abstract = "Various kinds of heart diseases pose a serious threat to human health. To effectively treat and prevent these diseases, accurate segmentation of the entire heart structure is crucial for medical research and application. At present, the solution to this problem still needs to rely on a lot of manpower. Not only is this time-consuming, but accuracy is sometimes difficult to guarantee. In the deep learning methods for medical image segmentation, large labeled images are difficult to obtain. Typically, the large databases have several thousand images, of which only a few hundred have been annotated, and the number of individual patients is even smaller. In this article, we focus on a small part of the dataset to minimize the cost of manual labeling and maximize the accurate segmentation results. The small part of the dataset contains more representative and informative images, avoiding doctors to repeatedly label images with similar information. We proposed a balanced distribution active learning (BDAL) framework for MRI cardiac multistructures segmentation based on reinforcement learning. The deep Q-network framework can learn an effective policy to select some informative and representative images to be labeled from a large number of the unlabeled dataset. We consider the shape features of images and the balance of different class distributions to build new state and action representation, which can help the agent to identify informative and representative images for annotation. Our BDAL method provides an agent to improve the ability of AL to select images to improve the accuracy of segmentation. Moreover, experiments and results show that our BDAL method significantly outperforms all baselines and other AL-based methods under the same amount of annotation budget on MRI cardiac multistructures segmentation in datasets ACDC and M&Ms .",
author = "Xiu Shu and Yunyun Yang and Jun Liu and Xiaojun Chang and Boying Wu",
year = "2024",
month = apr,
day = "30",
doi = "10.1109/TII.2023.3342442",
language = "English",
volume = "20",
pages = "6099--6108",
journal = "IEEE Transactions on Industrial Informatics",
issn = "1551-3203",
publisher = "IEEE Computer Society",
number = "4",

}

RIS

TY - JOUR

T1 - BDAL

T2 - Balanced Distribution Active Learning for MRI Cardiac Multistructures Segmentation

AU - Shu, Xiu

AU - Yang, Yunyun

AU - Liu, Jun

AU - Chang, Xiaojun

AU - Wu, Boying

PY - 2024/4/30

Y1 - 2024/4/30

N2 - Various kinds of heart diseases pose a serious threat to human health. To effectively treat and prevent these diseases, accurate segmentation of the entire heart structure is crucial for medical research and application. At present, the solution to this problem still needs to rely on a lot of manpower. Not only is this time-consuming, but accuracy is sometimes difficult to guarantee. In the deep learning methods for medical image segmentation, large labeled images are difficult to obtain. Typically, the large databases have several thousand images, of which only a few hundred have been annotated, and the number of individual patients is even smaller. In this article, we focus on a small part of the dataset to minimize the cost of manual labeling and maximize the accurate segmentation results. The small part of the dataset contains more representative and informative images, avoiding doctors to repeatedly label images with similar information. We proposed a balanced distribution active learning (BDAL) framework for MRI cardiac multistructures segmentation based on reinforcement learning. The deep Q-network framework can learn an effective policy to select some informative and representative images to be labeled from a large number of the unlabeled dataset. We consider the shape features of images and the balance of different class distributions to build new state and action representation, which can help the agent to identify informative and representative images for annotation. Our BDAL method provides an agent to improve the ability of AL to select images to improve the accuracy of segmentation. Moreover, experiments and results show that our BDAL method significantly outperforms all baselines and other AL-based methods under the same amount of annotation budget on MRI cardiac multistructures segmentation in datasets ACDC and M&Ms .

AB - Various kinds of heart diseases pose a serious threat to human health. To effectively treat and prevent these diseases, accurate segmentation of the entire heart structure is crucial for medical research and application. At present, the solution to this problem still needs to rely on a lot of manpower. Not only is this time-consuming, but accuracy is sometimes difficult to guarantee. In the deep learning methods for medical image segmentation, large labeled images are difficult to obtain. Typically, the large databases have several thousand images, of which only a few hundred have been annotated, and the number of individual patients is even smaller. In this article, we focus on a small part of the dataset to minimize the cost of manual labeling and maximize the accurate segmentation results. The small part of the dataset contains more representative and informative images, avoiding doctors to repeatedly label images with similar information. We proposed a balanced distribution active learning (BDAL) framework for MRI cardiac multistructures segmentation based on reinforcement learning. The deep Q-network framework can learn an effective policy to select some informative and representative images to be labeled from a large number of the unlabeled dataset. We consider the shape features of images and the balance of different class distributions to build new state and action representation, which can help the agent to identify informative and representative images for annotation. Our BDAL method provides an agent to improve the ability of AL to select images to improve the accuracy of segmentation. Moreover, experiments and results show that our BDAL method significantly outperforms all baselines and other AL-based methods under the same amount of annotation budget on MRI cardiac multistructures segmentation in datasets ACDC and M&Ms .

U2 - 10.1109/TII.2023.3342442

DO - 10.1109/TII.2023.3342442

M3 - Journal article

VL - 20

SP - 6099

EP - 6108

JO - IEEE Transactions on Industrial Informatics

JF - IEEE Transactions on Industrial Informatics

SN - 1551-3203

IS - 4

ER -