no code implementations • 14 May 2024 • Shishuai Hu, Zehui Liao, Zeyou Liu, Yong Xia
Extensive experiments on a public dataset underscore the superiority of our HiTTA over existing TTA methods, emphasizing the advantages of integrating human feedback and our divergence loss in enhancing the model's performance and adaptability across diverse medical centers.
2 code implementations • 17 Aug 2023 • Xianze Ai, Zehui Liao, Yong Xia
Although researchers adopt the label-noise-robust methods to handle label noise for lung nodule malignancy grading, they do not consider the inherent ordinal relation among classes of this task.
1 code implementation • 8 Jun 2023 • Shishuai Hu, Zehui Liao, Yong Xia
In C$^2$SDG, the shallower features of each image and its style-augmented counterpart are extracted and used for contrastive training, resulting in the disentangled style representations and structure representations.
no code implementations • 2 Jun 2023 • Zehui Liao, Yutong Xie, Shishuai Hu, Yong Xia
This paper proposes a Transformer-based Annotation Bias-aware (TAB) medical image segmentation model, which tackles the annotator-related bias via modeling annotator preference and stochastic errors.
no code implementations • 16 Dec 2022 • Zehui Liao, Shishuai Hu, Yutong Xie, Yong Xia
Specifically, we estimate the noisy posterior under the supervision of noisy labels, and approximate the batch-level noise transition matrix by estimating the inter-class correlation matrix with neither anchor points nor pseudo anchor points.
1 code implementation • 21 Nov 2022 • Shishuai Hu, Zehui Liao, Yong Xia
In this paper, we propose a \textbf{Pro}mpt learning based \textbf{SFDA} (\textbf{ProSFDA}) method for medical image segmentation, which aims to improve the quality of domain adaption by minimizing explicitly the domain discrepancy.
1 code implementation • 29 Aug 2022 • Shishuai Hu, Yiwen Ye, Zehui Liao, Yong Xia
Although numerous deep learning models have achieved remarkable success in many medical image segmentation tasks, accurate segmentation of kidney structures on computed tomography angiography (CTA) images remains challenging, due to the variable sizes of kidney tumors and the ambiguous boundaries between kidney structures and their surroundings.
1 code implementation • 29 Aug 2022 • Shishuai Hu, Zehui Liao, Yong Xia
In this paper, we propose a boundary-aware network (BA-Net) to segment abdominal organs on CT scans and MRI scans.
1 code implementation • 29 Aug 2022 • Shishuai Hu, Zehui Liao, Yong Xia
Carotid vessel wall segmentation is a crucial yet challenging task in the computer-aided diagnosis of atherosclerosis.
1 code implementation • 26 Nov 2021 • Zehui Liao, Shishuai Hu, Yutong Xie, Yong Xia
Manual annotation of medical images is highly subjective, leading to inevitable and huge annotation biases.
1 code implementation • 13 Sep 2021 • Shishuai Hu, Zehui Liao, Jianpeng Zhang, Yong Xia
In the DAC module, a dynamic convolutional head is conditioned on the predicted domain code of the input to make our model adapt to the unseen target domain.
1 code implementation • 23 Apr 2021 • Zehui Liao, Yutong Xie, Shishuai Hu, Yong Xia
According to the consistency and reliability of their annotations, we divide nodules into three sets: a consistent and reliable set (CR-Set), an inconsistent set (IC-Set), and a low reliable set (LR-Set).
no code implementations • 25 Nov 2020 • Yutong Xie, Jianpeng Zhang, Zehui Liao, Yong Xia, Chunhua Shen
In this paper, we propose a PriorGuided Local (PGL) self-supervised model that learns the region-wise local consistency in the latent feature space.