Semi-supervised Medical Image Segmentation through Dual-task Consistency

Luo, Xiangde; Chen, Jieneng; Song, Tao; Chen, Yinan; Guotai, Wang; Zhang, Shaoting

doi:10.1609/aaai.v35i10.17066

Cited by 339 publications

(106 citation statements)

References 42 publications

Supporting

Mentioning

106

Contrasting

Order By: Relevance

“…When using the MSD dataset as the source domain, the performance of pretrained model directly testing on the NIH dataset yielded 69.46%, while the results after proposed adaptation could achieve 77.13% (with a gain of 7.67%). Luo et al 47 . proposed a semisupervised segmentation method using a portion of labeled data from NIH dataset to train the segmentation model and obtained a DSC score of 78.27%.…”

Section: Resultsmentioning

confidence: 99%

Multiscale unsupervised domain adaptation for automatic pancreas segmentation in CT volumes using adversarial learning

Zhu

et al. 2022

Medical Physics

View full text Add to dashboard Cite

Computer-aided automatic pancreas segmentation is essential for early diagnosis and treatment of pancreatic diseases. However, the annotation of pancreas images requires professional doctors and considerable expenditure. Due to imaging differences among various institution population, scanning devices, imaging protocols, and so on, significant degradation in the performance of model inference results is prone to occur when models trained with domain-specific (usually institution-specific) datasets are directly applied to new (other centers/institutions) domain data. In this paper, we propose a novel unsupervised domain adaptation method based on adversarial learning to address pancreas segmentation challenges with the lack of annotations and domain shift interference. Methods: A 3D semantic segmentation model with attention module and residual module is designed as the backbone pancreas segmentation model. In both segmentation model and domain adaptation discriminator network, a multiscale progressively weighted structure is introduced to acquire different field of views. Features of labeled data and unlabeled data are fed in pairs into the proposed multiscale discriminator to learn domain-specific characteristics. Then the unlabeled data features with pseudodomain label are fed to the discriminator to acquire domain-ambiguous information. With this adversarial learning strategy, the performance of the segmentation network is enhanced to segment unseen unlabeled data. Results: Experiments were conducted on two public annotated datasets as source datasets, respectively, and one private dataset as target dataset, where annotations were not used for the training process but only for evaluation. The 3D segmentation model achieves comparative performance with state-of -theart pancreas segmentation methods on source domain. After implementing our domain adaptation architecture, the average dice similarity coefficient (DSC) of the segmentation model trained on the NIH-TCIA source dataset increases from 58.79% to 72.73% on the local hospital dataset, while the performance of the target domain segmentation model transferred from the medical segmentation decathlon (MSD) source dataset rises from 62.34% to 71.17%. Conclusions: Correlations of features across data domains are utilized to train the pancreas segmentation model on unlabeled data domain, improving the generalization of the model. Our results demonstrate that the proposed method enables the segmentation model to make meaningful segmentation for unseen data of the training set. In the future, the proposed method has the potential to

show abstract

Section: Resultsmentioning

confidence: 99%

Multiscale unsupervised domain adaptation for automatic pancreas segmentation in CT volumes using adversarial learning

Zhu

et al. 2022

Medical Physics

View full text Add to dashboard Cite

show abstract

“…For example, Fagherazzi, et al [ 137 ] presented a semi-supervised clustering technique for healthcare data, while Yu, et al [ 138 ] suggested a semi-supervised ML strategy for activity detection using sensor data. Peng, et al [ 139 ] and Luo, et al [ 140 ] used a semi-supervised learning strategy to segment medical images.…”

Section: Artificial Intelligencementioning

confidence: 99%

Recent Advancements in Emerging Technologies for Healthcare Management Systems: A Survey

et al. 2022

View full text Add to dashboard Cite

In recent times, the growth of the Internet of Things (IoT), artificial intelligence (AI), and Blockchain technologies have quickly gained pace as a new study niche in numerous collegiate and industrial sectors, notably in the healthcare sector. Recent advancements in healthcare delivery have given many patients access to advanced personalized healthcare, which has improved their well-being. The subsequent phase in healthcare is to seamlessly consolidate these emerging technologies such as IoT-assisted wearable sensor devices, AI, and Blockchain collectively. Surprisingly, owing to the rapid use of smart wearable sensors, IoT and AI-enabled technology are shifting healthcare from a conventional hub-based system to a more personalized healthcare management system (HMS). However, implementing smart sensors, advanced IoT, AI, and Blockchain technologies synchronously in HMS remains a significant challenge. Prominent and reoccurring issues such as scarcity of cost-effective and accurate smart medical sensors, unstandardized IoT system architectures, heterogeneity of connected wearable devices, the multidimensionality of data generated, and high demand for interoperability are vivid problems affecting the advancement of HMS. Hence, this survey paper presents a detailed evaluation of the application of these emerging technologies (Smart Sensor, IoT, AI, Blockchain) in HMS to better understand the progress thus far. Specifically, current studies and findings on the deployment of these emerging technologies in healthcare are investigated, as well as key enabling factors, noteworthy use cases, and successful deployments. This survey also examined essential issues that are frequently encountered by IoT-assisted wearable sensor systems, AI, and Blockchain, as well as the critical concerns that must be addressed to enhance the application of these emerging technologies in the HMS.

show abstract

“…Adversarial learningbased methods [15] use discriminators to encourage the predictions of unannotated images to be similar to those of the annotated ones, where prior information [18] may be used to improve the performance. Consistency regularization methods encourage predictions from one input image under different perturbations to be consistent, e.g., transformation consistency [19], dual-task consistency [20], and perturbationbased consistency [21].…”

Section: A Semi-supervised Learningmentioning

confidence: 99%

Contrastive Semi-supervised Learning for Domain Adaptive Segmentation Across Similar Anatomical Structures

Gu¹,

Zhang²,

Wang³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance for medical image segmentation, yet need plenty of manual annotations for training. Semi-Supervised Learning (SSL) methods are promising to reduce the requirement of annotations, but their performance is still limited when the dataset size and the number of annotated images are small. Leveraging existing annotated datasets with similar anatomical structures to assist training has a potential for improving the model's performance. However, it is further challenged by the cross-anatomy domain shift due to the different appearance and even imaging modalities from the target structure. To solve this problem, we propose Contrastive Semisupervised learning for Cross Anatomy Domain Adaptation (CS-CADA) that adapts a model to segment similar structures in a target domain, which requires only limited annotations in the target domain by leveraging a set of existing annotated images of similar structures in a source domain. We use Domain-Specific Batch Normalization (DSBN) to individually normalize feature maps for the two anatomical domains, and propose a cross-domain contrastive learning strategy to encourage extracting domain invariant features. They are integrated into a Self-Ensembling Mean-Teacher (SE-MT) framework to exploit unlabeled target domain images with a prediction consistency constraint. Extensive experiments show that our CS-CADA is able to solve the challenging cross-anatomy domain shift problem, achieving accurate segmentation of coronary arteries in X-ray images with the help of retinal vessel images and cardiac MR images with the help of fundus images, respectively, given only a small number of annotations in the target domain.

show abstract

Semi-supervised Medical Image Segmentation through Dual-task Consistency

Cited by 339 publications

References 42 publications

Multiscale unsupervised domain adaptation for automatic pancreas segmentation in CT volumes using adversarial learning

Multiscale unsupervised domain adaptation for automatic pancreas segmentation in CT volumes using adversarial learning

Recent Advancements in Emerging Technologies for Healthcare Management Systems: A Survey

Contrastive Semi-supervised Learning for Domain Adaptive Segmentation Across Similar Anatomical Structures

Contact Info

Product

Resources

About