TY - GEN
T1 - Punctate White Matter Lesion Segmentation in Preterm Infants Powered by Counterfactually Generative Learning
AU - Ren, Zehua
AU - Sun, Yongheng
AU - Wang, Miaomiao
AU - Feng, Yuying
AU - Li, Xianjun
AU - Jin, Chao
AU - Yang, Jian
AU - Lian, Chunfeng
AU - Wang, Fan
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
PY - 2023
Y1 - 2023
N2 - Accurate segmentation of punctate white matter lesions (PWMLs) are fundamental for the timely diagnosis and treatment of related developmental disorders. Automated PWMLs segmentation from infant brain MR images is challenging, considering that the lesions are typically small and low-contrast, and the number of lesions may dramatically change across subjects. Existing learning-based methods directly apply general network architectures to this challenging task, which may fail to capture detailed positional information of PWMLs, potentially leading to severe under-segmentations. In this paper, we propose to leverage the idea of counterfactual reasoning coupled with the auxiliary task of brain tissue segmentation to learn fine-grained positional and morphological representations of PWMLs for accurate localization and segmentation. A simple and easy-to-implement deep-learning framework (i.e., DeepPWML) is accordingly designed. It combines the lesion counterfactual map with the tissue probability map to train a lightweight PWML segmentation network, demonstrating state-of-the-art performance on a real-clinical dataset of infant T1w MR images. The code is available at https://github.com/ladderlab-xjtu/DeepPWML.
AB - Accurate segmentation of punctate white matter lesions (PWMLs) are fundamental for the timely diagnosis and treatment of related developmental disorders. Automated PWMLs segmentation from infant brain MR images is challenging, considering that the lesions are typically small and low-contrast, and the number of lesions may dramatically change across subjects. Existing learning-based methods directly apply general network architectures to this challenging task, which may fail to capture detailed positional information of PWMLs, potentially leading to severe under-segmentations. In this paper, we propose to leverage the idea of counterfactual reasoning coupled with the auxiliary task of brain tissue segmentation to learn fine-grained positional and morphological representations of PWMLs for accurate localization and segmentation. A simple and easy-to-implement deep-learning framework (i.e., DeepPWML) is accordingly designed. It combines the lesion counterfactual map with the tissue probability map to train a lightweight PWML segmentation network, demonstrating state-of-the-art performance on a real-clinical dataset of infant T1w MR images. The code is available at https://github.com/ladderlab-xjtu/DeepPWML.
UR - https://www.scopus.com/pages/publications/85174698331
U2 - 10.1007/978-3-031-43904-9_22
DO - 10.1007/978-3-031-43904-9_22
M3 - 会议稿件
AN - SCOPUS:85174698331
SN - 9783031439032
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 220
EP - 229
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2023 - 26th International Conference, Proceedings
A2 - Greenspan, Hayit
A2 - Greenspan, Hayit
A2 - Madabhushi, Anant
A2 - Mousavi, Parvin
A2 - Salcudean, Septimiu
A2 - Duncan, James
A2 - Syeda-Mahmood, Tanveer
A2 - Taylor, Russell
PB - Springer Science and Business Media Deutschland GmbH
T2 - 26th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2023
Y2 - 8 October 2023 through 12 October 2023
ER -