Domain Shift Research Articles

A domain generalization network for imbalanced machinery fault diagnosis

Traditional models for Imbalanced Fault Diagnosis (IFD) face challenges in practical applications due to domain shifts caused by varying working conditions and machinery. Domain Generalization (DG) models provide an advantage over traditional approaches by learning class-discriminative and domain-invariant feature representations, allowing them to generalize to unseen target data. However, the scarcity of fault samples relative to healthy ones limits their application in real-world industrial scenarios. In this paper, we propose a Domain Mixed-Enhanced Domain Generalization Network (DEMDGN) that enhances IFD performance by utilizing mixup-based data augmentation and domain-based discrepancy metrics to align feature distributions across multiple heterogeneous source domains. By creating domain-invariant features, DEMDGN allows robust fault diagnosis under varying conditions. Extensive experiments on one marine machinery dataset and two bearing datasets demonstrate that the proposed method effectively addresses class imbalance and domain shift problems, achieving superior diagnostic performance.

Adversarial Unsupervised Domain Adaptation for 3D Semantic Segmentation with 2D Image Fusion of Dense Depth

AbstractUnsupervised domain adaptation (UDA) is increasingly used for 3D point cloud semantic segmentation tasks due to its ability to address the issue of missing labels for new domains. However, most existing unsupervised domain adaptation methods focus only on uni‐modal data and are rarely applied to multi‐modal data. Therefore, we propose a cross‐modal UDA on multi‐modal datasets that contain 3D point clouds and 2D images for 3D Semantic Segmentation. Specifically, we first propose a Dual discriminator‐based Domain Adaptation (Dd‐bDA) module to enhance the adaptability of different domains. Second, given that the robustness of depth information to domain shifts can provide more details for semantic segmentation, we further employ a Dense depth Feature Fusion (DdFF) module to extract image features with rich depth cues. We evaluate our model in four unsupervised domain adaptation scenarios, i.e., dataset‐to‐dataset (A2D2 → SemanticKITTI), Day‐to‐Night, country‐to‐country (USA → Singapore), and synthetic‐to‐real (VirtualKITTI → SemanticKITTI). In all settings, the experimental results achieve significant improvements and surpass state‐of‐the‐art models.

An effective unsupervised domain adaptation for in-field potato disease recognition

Accurate disease recognition through computer vision is crucial for the intelligent management of potato production. Popular data-driven classification methods face challenges including limited labelled data and poor model portability. Unsupervised Domain Adaptation (UDA) addresses these challenges with a novel learning strategy. However, the complex field environment introduces a significant domain shift problem due to varying conditions. Existing UDA methods usually concentrate on aligning global data distribution and employ a single structure for disease feature extraction, thereby limiting their efficacy in true field environment. To tackle this challenge of potato disease recognition, the Multi-Representation Adaptive Network (MRSAN) based on subdomain alignment is presented. MRSAN effectively aligns feature distributions across diverse data by minimising distribution differences among relevant subdomains. Simultaneously, the multi-representation extraction method captures finer details from various perspectives in the disease images. The combination of these two approaches efficiently mitigates the adverse effects caused by various interference factors in field environment. Based on the acquisition conditions of light variation and disease progression, two field potato disease image datasets are created, containing five and six kinds of potato leaf disease, respectively. Extensive transfer experiments are conducted on the two datasets. MRSAN achieves average classification accuracies of 87.03% and 80.06% on the datasets for the corresponding transfer tasks, outperforming the other compared methods. This not only validates the effectiveness of MRSAN but also demonstrates its robust ability to generalise across changes in regard to light variation and disease progression.

Artificial intelligence can be trained to predict c-KIT-11 mutational status of canine mast cell tumors from hematoxylin and eosin-stained histological slides.

Numerous prognostic factors are currently assessed histologically and immunohistochemically in canine mast cell tumors (MCTs) to evaluate clinical behavior. In addition, polymerase chain reaction (PCR) is often performed to detect internal tandem duplication (ITD) mutations in exon 11 of the c-KIT gene (c-KIT-11-ITD) to predict the therapeutic response to tyrosine kinase inhibitors. This project aimed at training deep learning models (DLMs) to identify MCTs with c-KIT-11-ITD solely based on morphology. Hematoxylin and eosin (HE) stained slides of 368 cutaneous, subcutaneous, and mucocutaneous MCTs (195 with ITD and 173 without) were stained consecutively in 2 different laboratories and scanned with 3 different slide scanners. This resulted in 6 data sets (stain-scanner variations representing diagnostic institutions) of whole-slide images. DLMs were trained with single and mixed data sets and their performances were assessed under stain-scanner variations (domain shifts). The DLM correctly classified HE slides according to their c-KIT-11-ITD status in up to 87% of cases with a 0.90 sensitivity and a 0.83 specificity. A relevant performance drop could be observed when the stain-scanner combination of training and test data set differed. Multi-institutional data sets improved the average accuracy but did not reach the maximum accuracy of algorithms trained and tested on the same stain-scanner variant (ie, intra-institutional). In summary, DLM-based morphological examination can predict c-KIT-11-ITD with high accuracy in canine MCTs in HE slides. However, staining protocol and scanner type influence accuracy. Larger data sets of scans from different laboratories and scanners may lead to more robust DLMs to identify c-KIT mutations in HE slides.

Domain adaptation for semantic segmentation of road scenes via two-stage alignment of traffic elements

Unsupervised domain adaptation has been used to reduce the domain shift, which would improve the performance of semantic segmentation on unlabeled real-world data. However, existing methodologies fall short in effectively addressing the domain shift issue prevalent in traffic scenarios, leading to less than satisfactory segmentation results. In this paper, we propose a novel domain adaptation method for semantic segmentation via unsupervised alignment of traffic elements. Firstly, we introduce a two-stage self-training framework that leverages a blended set of training samples to enhance the training process. In the first stage, we leverage generated mixup training samples as inputs within our two-stage self-training framework and have developed corresponding loss functions for both the source and target domains to direct the training process. Then, the alignment modules for dynamic and static traffic elements are designed to achieve accurate matching between the source and the target domain images. The cosine similarity maximization is applied to the alignment of dynamic traffic elements, while the prototype learning is utilized for the static traffic elements. Additionally, we present a new technique for reducing noise in pseudo labels by constructing thresholds that adjust to each class. Meanwhile, we formulate the associated target domain loss function for vacant pseudo label pixels. The experimental results demonstrate that the proposed method is superior to the existing methods on five different domain adaptation tasks, which is more applicable to semantic segmentation of road scenes.

Core-Periphery Multi-Modality Feature Alignment for Zero-Shot Medical Image Analysis.

Multi-modality learning, exemplified by the language-image pair pre-trained CLIP model, has demonstrated remarkable performance in enhancing zero-shot capabilities and has gained significant attention recently. However, simply applying language-image pre-trained CLIP to medical image analysis encounters substantial domain shifts, resulting in severe performance degradation due to inherent disparities between natural (non-medical) and medical image characteristics. To address this challenge and uphold or even enhance CLIP's zero-shot capability in medical image analysis, we develop a novel approach, Core-Periphery feature alignment for CLIP (CP-CLIP), to model medical images and corresponding clinical text jointly. To achieve this, we design an auxiliary neural network whose structure is organized by the core-periphery (CP) principle. This auxiliary CP network not only aligns medical image and text features into a unified latent space more efficiently but also ensures alignment driven by principles of brain network organization. In this way, our approach effectively mitigates and further enhances CLIP's zero-shot performance in medical image analysis. More importantly, the proposed CP-CLIP exhibits excellent explanatory capability, enabling the automatic identification of critical disease-related regions in clinical analysis. Extensive experiments and evaluation across five public datasets covering different diseases underscore the superiority of our CP-CLIP in zero-shot medical image prediction and critical features detection, showing its promising utility in multimodal feature alignment in current medical applications.

Multiexposed Image-Fusion Strategy Using Mutual Image Translation Learning with Multiscale Surround Switching Maps

The dynamic range of an image represents the difference between its darkest and brightest areas, a crucial concept in digital image processing and computer vision. Despite display technology advancements, replicating the broad dynamic range of the human visual system remains challenging, necessitating high dynamic range (HDR) synthesis, combining multiple low dynamic range images captured at contrasting exposure levels to generate a single HDR image that integrates the optimal exposure regions. Recent deep learning advancements have introduced innovative approaches to HDR generation, with the cycle-consistent generative adversarial network (CycleGAN) gaining attention due to its robustness against domain shifts and ability to preserve content style while enhancing image quality. However, traditional CycleGAN methods often rely on unpaired datasets, limiting their capacity for detail preservation. This study proposes an improved model by incorporating a switching map (SMap) as an additional channel in the CycleGAN generator using paired datasets. The SMap focuses on essential regions, guiding weighted learning to minimize the loss of detail during synthesis. Using translated images to estimate the middle exposure integrates these images into HDR synthesis, reducing unnatural transitions and halo artifacts that could occur at boundaries between various exposures. The multilayered application of the retinex algorithm captures exposure variations, achieving natural and detailed tone mapping. The proposed mutual image translation module extends CycleGAN, demonstrating superior performance in multiexposure fusion and image translation, significantly enhancing HDR image quality. The image quality evaluation indices used are CPBDM, JNBM, LPC-SI, S3, JPEG_2000, and SSEQ, and the proposed model exhibits superior performance compared to existing methods, recording average scores of 0.6196, 15.4142, 0.9642, 0.2838, 80.239, and 25.054, respectively. Therefore, based on qualitative and quantitative results, this study demonstrates the superiority of the proposed model.

Instance-wise multi-view visual fusion for zero-shot learning

Zero-shot learning (ZSL) has become increasing popular in computer vision due to its ability to recognize categories unobserved in the training data. So far, most existing ZSL approaches adopt visual representations that are either derived from pretrained networks or learned using an end-to-end architecture. However, a single group of visual representations can hardly capture all features hidden in the images, yielding incomplete visual information. In numerous real-life scenarios, multi-view visual representations are often accessible which describe the instances more comprehensively and are potential for better learning performance. In this paper, we introduce an instance-wise multi-view visual fusion (IMVF) for zero-shot learning (ZSL) model. In accordance with the consensus principle, a multi-view visual-semantic mapping is created by minimizing the disparities of seen-class semantic projections on different views. Meanwhile, a straightforward linear constraint is performed on each seen-class instance to adhere to the complementary principle so that the cross-view information exchange is well motivated. In order to mitigate the domain shift problem, the predicted unseen-class semantic projections are further refined through a multi-view manifold alignment under the consensus principle. Our proposed IMVFZSL is compared with the State-of-the-Art ZSL methods on AwA2, CUB and SUN datasets. Exciting experimental results validate the effectiveness of the IMVF mechanism. To the best of our understanding, this is an initial attempt to fuse multi-view visual representations in ZSL, which will stimulate valuable contemplation in this field.

Addressing the Generalizability of AI in Radiology Using a Novel Data Augmentation Framework with Synthetic Patient Image Data: Proof-of-Concept and External Validation Classification Tasks in Multiple Sclerosis.

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Artificial intelligence (AI) models often face performance drops after deployment to external datasets. This study evaluated the potential of a novel data augmentation framework based on generative adversarial networks (GAN) that creates synthetic patient image data during model training to improve model generalizability. Model development and external testing were performed for a given classification task, namely the detection of new fluid-attenuated inversion recovery (FLAIR) lesions on MRI during longitudinal follow-up of patients with multiple sclerosis (MS). An internal dataset of 669 patients with MS (n = 3083 examinations) was used to develop an attention-based network, trained both with and without the inclusion of the GAN-based synthetic data augmentation framework. External testing was performed on 134 patients with MS from a different institution, with MR images acquired using different scanners and protocols than images used during training. Models trained using synthetic data augmentation showed a significant performance improvement when applied on external data (AUC 83.6% without synthetic data versus AUC 93.3% with synthetic data augmentation, P = .03), achieving comparable results to the internal test set (AUC 95.5%, P = .53), whereas models without synthetic data augmentation demonstrated a performance drop upon external testing (AUC 93.8% on internal dataset versus AUC 83.6% on external data, P = .03). Data augmentation with synthetic patient data substantially improved performance of AI models on unseen MRI data and may be extended to other clinical conditions or tasks to mitigate domain shift, limit class imbalance, and enhance the robustness of AI applications in medical imaging. ©RSNA, 2024.

Domain knowledge-powered attention for air traffic management hazardous events classification

The automated classification of hazardous events in Air Traffic Management (ATM) involves concise texts with specialized terminology. The problem is currently addressed by machine learning text classification methods; however, some methods overlook domain knowledge, while others require it excessively. This research aims to identify a process for integrating low-cost, text-based domain knowledge into classification to address the issue of domain shift. To achieve this, the Wide and Deep Bidirectional Encoder Representations from Transformers (WD-BERT) model is proposed, featuring a unique knowledge-powered attention mechanism. WD-BERT's Wide Attention module assesses the compatibility between input text and domain knowledge to determine class probabilities, while its Deep Attention module extracts contextual features guided by domain knowledge. Additionally, a text mining method is employed to extract domain knowledge from ATM regulation texts. The model is trained and evaluated using a dataset derived from the China ATM Hazard Source Database. It achieves a multi-label classification accuracy of 80.24% and an F1-micro score of 0.88, outperforming comparative models with state-of-the-art performance. Breakdown analyses of the model indicate that the carefully designed attention mechanism endows WD-BERT with the ability to integrate domain knowledge, allowing it to excel in contexts involving complex sentences, multiple terms, and domain shifts. The innovation of this study lies in the proposed domain knowledge-powered attention mechanism, which allows simply organized domain knowledge to effectively guide ATM hazardous events classification, alleviating the impact of domain shift. This method can also be applied to other vertical domains without knowledge graphs.

Unpaired Image Translation to Mitigate Domain Shift in Liquid Argon Time Projection Chamber Detector Responses

Abstract Deep learning algorithms often are developed and trained on a training dataset and deployed on test datasets. Any systematic difference between the training and a test dataset may severely degrade the final algorithm performance on the test dataset -- what is known as the domain shift problem. This issue is prevalent in many scientific domains where algorithms are trained on simulated data but applied to real-world datasets. Typically, the domain shift problem is solved through various domain adaptation methods. However, these methods are often tailored to a specific downstream task and may not easily generalize to different tasks. This work explores the feasibility of using an alternative way to solve the domain shift problem that is not specific to any downstream algorithm. The proposed approach relies on modern Unpaired Image-to-Image translation techniques, designed to find translations between different image domains in a fully unsupervised fashion. In this study, the approach is applied to a domain shift problem commonly encountered in Liquid Argon Time Projection Chamber detector research when seeking a way to translate samples between two differently distributed LArTPC detector datasets deterministically. This translation allows for mapping real-world data into the simulated data domain where the downstream algorithms can be run with much less domain-shift-related performance degradation. Conversely, using the translation from the simulated data in a real-world domain can increase the realism of the simulated dataset and reduce the magnitude of any systematic uncertainties. We adapted several popular UI2I translation algorithms to work on scientific data and demonstrated the viability of these techniques for solving the domain shift problem with LArTPC detector data. To facilitate further development of domain adaptation techniques for scientific datasets, the "Simple Liquid-Argon Track Samples" (SLATS) dataset used in this study also is published.

Test time adaptation for cross-domain sleep stage classification

Domain shift is a serious problem with existing cross-domain sleep stage classification methods. However, most existing domain adaptation methods require access to a large amount of source domain data containing patient information, which may lead to infringement of patient privacy. At the same time, existing methods require collecting all information in advance when processing target data, which cannot meet the situation where test data arrives in batches or in order in actual scenarios. To this end, we propose a test time adaptation framework based on knowledge distillation for cross-domain sleep stage classification. This teacher-student network framework uses pre-trained model to learn variant features and robust representations. We evaluate our framework in the cross domain scenario composed of three datasets and validate its superiority, real time performance and privacy protection capability.

Multiresolution cascaded attention U-Net for localization and segmentation of optic disc and fovea in fundus images

Identification of retinal diseases in automated screening methods, such as those used in clinical settings or computer-aided diagnosis, usually depends on the localization and segmentation of the Optic Disc (OD) and fovea. However, this task is difficult since these anatomical features have irregular spatial, texture, and shape characteristics, limited sample sizes, and domain shifts due to different data distributions across datasets. This study proposes a novel Multiresolution Cascaded Attention U-Net (MCAU-Net) model that addresses these problems by optimally balancing receptive field size and computational efficiency. The MCAU-Net utilizes two skip connections to accurately localize and segment the OD and fovea in fundus images. We incorporated a Multiresolution Wavelet Pooling Module (MWPM) into the CNN at each stage of U-Net input to compensate for spatial information loss. Additionally, we integrated a cascaded connection of the spatial and channel attentions as a skip connection in MCAU-Net to concentrate precisely on the target object and improve model convergence for segmenting and localizing OD and fovea centers. The proposed model has a low parameter count of 0.8 million, improving computational efficiency and reducing the risk of overfitting. For OD segmentation, the MCAU-Net achieves high IoU values of 0.9771, 0.945, and 0.946 for the DRISHTI-GS, DRIONS-DB, and IDRiD datasets, respectively, outperforming previous results for all three datasets. For the IDRiD dataset, the MCAU-Net locates the OD center with an Euclidean Distance (ED) of 16.90 pixels and the fovea center with an ED of 33.45 pixels, demonstrating its effectiveness in overcoming the common limitations of state-of-the-art methods.

Acoustic insights into the corn extrusion process for enhanced quality control

In industrial extrusion processes, a solid material is pressed through a die to obtain products of the desired shape and dimension. Fluctuations in the process parameters have a significant impact on the product quality. In food extrusion, the expansion noise at the die can serve as an indicator of the stability of the process. This study employs microphones to characterize the corn extrusion process, focusing on correlating acoustic emissions with predefined process parameters such as feed intake and water content. Experimental data from laboratory and industrial settings reveal distinct domain shifts, yet consistent findings confirm the distinguishability of various extrusion process parameters by analyzing acoustic emissions. Employing machine learning models, including support vector machines and convolutional neural networks, in conjunction with audio features such as log Mel spectrograms, yields promising accuracies above 90\% in discriminating between standard and non-standard process parameters. The proposed acoustic quality control approach has the potential to enhance the stability of extrusion processes and contributes to the development of automated monitoring systems in the field of food extrusion. This ensures consistent quality and reduced waste, ultimately leading to significant cost savings in production.

Hierarchical Keypoints Feature Alignment for Domain Adaptive Pose Estimation

Unsupervised Domain Adaptation (UDA) is a popular research topic in computer vision. One of the most common strategies in this field is to reduce domain shifts through global feature alignment. Unfortunately, we observe its failure on adaptive human pose estimation. From our analysis, we find out two major reasons: the extreme imbalance between keypoints and non-keypoints regions and the high diversity of human skeleton structures. To address these problems, we propose a simple yet effective approach named PoseDA. Our idea is to let the alignment focus on the feature of keypoint regions and enforce the model to learn domain-invariant representations for keypoint prediction. The key component of our PoseDA is a Hierarchical Feature Selection of Keypoints Regions (HFS) module, which consists of Coarse Feature Selection of Keypoints Regions (CFS) and Reliable Feature Selection of Keypoints Regions (RFS). By using HFS, we obtain reliable and compact keypoints features, allowing our model to achieve more effective feature alignment. Under three UDA scenarios, i.e. JTA→MPII, SURREAL→LSP, SURREAL→Human3.6, our PoseDA establishes new state-of-the-art performance. In particular, our PoseDA outperforms the previous best UDA methods by over 7% w.r.t. PCKh on JTA→MPII.

Learn Then Adapt: A Novel Test-Time Adaptation Method for Cross-Domain Fault Diagnosis of Rolling Bearings

Cross-domain fault diagnosis enhances the generalization capability of diagnostic models across different operating conditions and machines. Current studies tackle the domain shift problem by adapting the model during training with data from the target domain or multiple source domains. However, a more realistic and less explored scenario is automatically adapting a trained (developed) model at test time (deployment period) using limited normal-condition data. To bridge this research gap, we propose a novel test-time adaptation framework to rapidly and effectively adapt the trained model, which only requires mini-batch test data (normal condition). Specifically, we first transform input signals to informative signal embedding and mitigate its noise with a reconstruction loss. Then, we decompose the signal embedding to the domain-related healthy component and the domain-invariant faulty component to better leverage the normal-condition data. Finally, we adapt the model by re-identifying the normal signals of the target domain during the test stage. Extensive experiments verify the effectiveness of our method, demonstrating performance improvements across public and private datasets.

A Patch Diversity Transformer for Domain Generalized Semantic Segmentation.

Domain generalization (DG) is one of the critical issues for deep learning in unknown domains. How to effectively represent domain-invariant context (DIC) is a difficult problem that DG needs to solve. Transformers have shown the potential to learn generalized features, since the powerful ability to learn global context. In this article, a novel method named patch diversity Transformer (PDTrans) is proposed to improve the DG for scene segmentation by learning global multidomain semantic relations. Specifically, patch photometric perturbation (PPP) is proposed to improve the representation of multidomain in the global context information, which helps the Transformer learn the relationship between multiple domains. Besides, patch statistics perturbation (PSP) is proposed to model the feature statistics of patches under different domain shifts, which enables the model to encode domain-invariant semantic features and improve generalization. PPP and PSP can help to diversify the source domain at the patch level and feature level. PDTrans learns context across diverse patches and takes advantage of self-attention to improve DG. Extensive experiments demonstrate the tremendous performance advantages of the PDTrans over state-of-the-art DG methods.

Reducing semantic ambiguity in domain adaptive semantic segmentation via probabilistic prototypical pixel contrast

Domain adaptation aims to reduce the model degradation on the target domain caused by the domain shift between the source and target domains. Although encouraging performance has been achieved by combining contrastive learning with the self-training paradigm, they suffer from ambiguous scenarios caused by scale, illumination, or overlapping when deploying deterministic embedding. To address these issues, we propose probabilistic prototypical pixel contrast (PPPC), a universal adaptation framework that models each pixel embedding as a probability via multivariate Gaussian distribution to fully exploit the uncertainty within them, eventually improving the representation quality of the model. In addition, we derive prototypes from probability estimation posterior probability estimation which helps to push the decision boundary away from the ambiguity points. Moreover, we employ an efficient method to compute similarity between distributions, eliminating the need for sampling and reparameterization, thereby significantly reducing computational overhead. Further, we dynamically select the ambiguous crops at the image level to enlarge the number of boundary points involved in contrastive learning, which benefits the establishment of precise distributions for each category. Extensive experimentation demonstrates that PPPC not only helps to address ambiguity at the pixel level, yielding discriminative representations but also achieves significant improvements in both synthetic-to-real and day-to-night adaptation tasks. It surpasses the previous state-of-the-art (SOTA) by +5.2% mIoU in the most challenging daytime-to-nighttime adaptation scenario, exhibiting stronger generalization on other unseen datasets. The code and models are available at https://github.com/DarlingInTheSV/Probabilistic-Prototypical-Pixel-Contrast.

A Foundation Language-Image Model of the Retina (FLAIR): encoding expert knowledge in text supervision

Foundation vision-language models are currently transforming computer vision, and are on the rise in medical imaging fueled by their very promising generalization capabilities. However, the initial attempts to transfer this new paradigm to medical imaging have shown less impressive performances than those observed in other domains, due to the significant domain shift and the complex, expert domain knowledge inherent to medical-imaging tasks. Motivated by the need for domain-expert foundation models, we present FLAIR, a pre-trained vision-language model for universal retinal fundus image understanding. To this end, we compiled 38 open-access, mostly categorical fundus imaging datasets from various sources, with up to 101 different target conditions and 288,307 images. We integrate the expert’s domain knowledge in the form of descriptive textual prompts, during both pre-training and zero-shot inference, enhancing the less-informative categorical supervision of the data. Such a textual expert’s knowledge, which we compiled from the relevant clinical literature and community standards, describes the fine-grained features of the pathologies as well as the hierarchies and dependencies between them. We report comprehensive evaluations, which illustrate the benefit of integrating expert knowledge and the strong generalization capabilities of FLAIR under difficult scenarios with domain shifts or unseen categories. When adapted with a lightweight linear probe, FLAIR outperforms fully-trained, dataset-focused models, more so in the few-shot regimes. Interestingly, FLAIR outperforms by a wide margin larger-scale generalist image-language models and retina domain-specific self-supervised networks, which emphasizes the potential of embedding experts’ domain knowledge and the limitations of generalist models in medical imaging. The pre-trained model is available at: https://github.com/jusiro/FLAIR.

Adaptive wavelet-VNet for single-sample test time adaptation in medical image segmentation.

In medical image segmentation, a domain gap often exists between training and testing datasets due to different scanners or imaging protocols, which leads to performance degradation in deep learning-based segmentation models. Given the high cost of manual labeling and the need for privacy protection, it is often challenging to annotate the testing (target) domain data for model fine-tuning or to collect data from different domains to train domain generalization models. Therefore, using only unlabeled target domain data for test-time adaptation (TTA) presents a more practical but challenging solution. To improve the segmentation accuracy of deep learning-based models on unseen datasets, and especially to enhance the efficiency and stability of TTA for individual samples from heterogeneous domains. In this study, we proposed to dynamically adapt a wavelet-VNet (WaVNet) to unseen target domains with a hybrid objective function, based on each unlabeled test sample during the test time. We embedded multiscale wavelet coefficients into a V-Net encoder and adaptively adjusted the spatial and spectral features according to the input, and the model parameters were optimized by three loss functions. We integrated a shape-aware loss to focus on the foreground segmentations, a Refine loss to correct the incomplete and noisy segmentations caused by domain shifts, and an entropy loss to promote the global consistency of the segmentations. We evaluated the proposed method on multidomain liver and prostate segmentation datasets to assess its advantages over other TTA methods. For the source domain model training of the liver dataset, we used 15 3D MR image samples for training and 5 for validation. Correspondingly, for the prostate dataset, we used 22 3D MR image samples for training and 7 for validation. In the target domain, we used a single 3D MR image sample for adaptation and testing. The total number of testing samples is 60 in the liver dataset (for 3 different domains) and 116 in the prostate dataset (for 6 different domains). The proposed method showed the highest segmentation accuracy among all methods, achieving a mean(±SD) Dice coefficient (DSC) of 78.10±5.23% and a mean 95th Hausdorff distance (HD95) of 15.52±5.84mm on the liver dataset; and a mean DSC of 80.02±3.89% and a mean HD95 of 9.18±3.47mm on the prostate dataset. The DSC is 11.67% (in absolute terms) and 15.27% higher than that of the baseline (no adaptation) method, for the liver and the prostate datasets, respectively. The proposed adaptive WaVNet enhanced the image segmentation accuracy from unseen domains during the test time via unsupervised learning and multi-objective optimization. It can benefit clinical applications where data are scarce or with changing data distributions, including online adaptive radiotherapy. The code will be released at: https://github.com/sanny1226/WaVNet.