Contrastive learning has shown great potential in unsupervised domain adaptation (UDA) for tackling annotation scarcity and domain shift in cross-modality medical image segmentation. However, most existing approaches rely on a single (global) prototype per class, derived from pseudo-labels at intermediate feature levels. This method is vulnerable to noise and fails to capture intra-class structural diversity, limiting the robustness of pixel-level semantic alignment. In this work, we propose a novel contrastive learning approach for unsupervised cross-modality medical image segmentation. Specifically, (1) we introduce a dual semantic anchor strategy that models both the core representation and structural diversity of each organ class by constructing a global anchor alongside multiple local anchors; (2) we design a tri-domain collaborative contrastive alignment mechanism that adaptively selects anchors based on data origin and uncertainty for fine-grained semantic alignment; (3) we propose a dynamic anchor consistency loss that alleviates early pseudo-label errors using a curriculum learning strategy, providing enhanced semantic constraints for the target domain. Our method significantly improves performance on two public medical image segmentation datasets, AMOS and PI-CAI. Extensive ablation studies confirm the effectiveness of each key component. The code can be accessed at https://github.com/caixukun11845/DATCA.

Domain adaptation techniques have been demonstrated to be effective in addressing label deficiency challenges in medical image segmentation. However, conventional domain adaptation based approaches often concentrate on matching global marginal distributions between different domains in a class-agnostic fashion. In this paper, we present a dual-attention domain-adaptative segmentation network (DADASeg-Net) for cross-modality medical image segmentation. The key contribution of DADASeg-Net is a novel dual adversarial attention mechanism, which regularizes the domain adaptation module with two attention maps respectively from the space and class perspectives. Specifically, the spatial attention map guides the domain adaptation module to focus on regions that are challenging to align in adaptation. The class attention map encourages the domain adaptation module to capture class-specific instead of class-agnostic knowledge for distribution alignment. DADASeg-Net shows superior performance in two challenging medical image segmentation tasks.

Synthesizing a frontal view face from a single nonfrontal image, i.e. face frontalization, is a task of practical importance in a wide range of facial image analysis applications. However, to train the frontalization model in a supervised manner, most existing face frontalization methods rely on the availability of nonfrontal-frontal face pairs (typically from the Multi-PIE dataset) captured in a constrained environment. Such approaches, in return, limit the generalizability of their application to unconstrained scenarios. Unfortunately, although a large amount of in-the-wild face datasets are available, they cannot easily be utilized for face frontalization training since the nonfrontal and frontal facial images are not paired. To train a frontalization network which generalizes well to both constrained and unconstrained environments, we propose a semi-supervised learning framework which effectively uses both (labeled) indoor and (unlabeled) outdoor faces. Specifically, to achieve this goal, this article presents a Cycle-Consistent Face Frontalization Generative Adversarial Network (CCFF-GAN) which consists of both (1) the supervised and (2) the unsupervised components. For (1), we use the indoor paired (labeled) data to learn a roughly accurate frontalization network which may not generalize well to outdoor (in-the-wild) scenarios. For (2), to cope with the generalization issue, the unsupervised part uses the unpaired (unlabeled) images under the perceptual cycle consistency constraint in the semantic feature space to generalize the network from controlled (indoor) to uncontrolled (outdoor) environment. Extensive experiments demonstrate the effectiveness of the proposed method in comparison with the state-of-the-art face frontalization methods, especially under the in-the-wild scenarios.

Orthognathic surgical outcomes rely heavily on the quality of surgical planning. Automatic estimation of a reference facial bone shape significantly reduces experience-dependent variability and improves planning accuracy and efficiency. We propose an end-to-end deep learning framework to estimate patient-specific reference bony shape models for patients with orthognathic deformities. Specifically, we apply a point-cloud network to learn a vertex-wise deformation field from a patient's deformed bony shape, represented as a point cloud. The estimated deformation field is then used to correct the deformed bony shape to output a patient-specific reference bony surface model. To train our network effectively, we introduce a simulation strategy to synthesize deformed bones from any given normal bone, producing a relatively large and diverse dataset of shapes for training. Our method was evaluated using both synthetic and real patient data. Experimental results show that our framework estimates realistic reference bony shape models for patients with varying deformities. The performance of our method is consistently better than an existing method and several deep point-cloud networks. Our end-to-end estimation framework based on geometric deep learning shows great potential for improving clinical workflows.

An increasing number of studies are leveraging unsupervised cross-modality synthesis to mitigate the limited label problem in training medical image segmentation models. They typically transfer ground truth annotations from a label-rich imaging modality to a label-lacking imaging modality, under an assumption that different modalities share the same anatomical structure information. However, since these methods commonly use voxel/pixel-wise cycle-consistency to regularize the mappings between modalities, high-level semantic information is not necessarily preserved. In this paper, we propose a novel anatomy-regularized representation learning approach for segmentation-oriented cross-modality image synthesis. It learns a common feature encoding across different modalities to form a shared latent space, where 1) the input and its synthesis present consistent anatomical structure information, and 2) the transformation between two images in one domain is preserved by their syntheses in another domain. We applied our method to the tasks of cross-modality skull segmentation and cardiac substructure segmentation. Experimental results demonstrate the superiority of our method in comparison with state-of-the-art cross-modality medical image segmentation methods.

Compared to computed tomography (CT), magnetic resonance imaging (MRI) delineation of craniomaxillofacial (CMF) bony structures can avoid harmful radiation exposure. However, bony boundaries are blurry in MRI, and structural information needs to be borrowed from CT during the training. This is challenging since paired MRI-CT data are typically scarce. In this paper, we propose to make full use of unpaired data, which are typically abundant, along with a single paired MRI-CT data to construct a one-shot generative adversarial model for automated MRI segmentation of CMF bony structures. Our model consists of a cross-modality image synthesis sub-network, which learns the mapping between CT and MRI, and an MRI segmentation sub-network. These two sub-networks are trained jointly in an end-to-end manner. Moreover, in the training phase, a neighbor-based anchoring method is proposed to reduce the ambiguity problem inherent in cross-modality synthesis, and a feature-matching-based semantic consistency constraint is proposed to encourage segmentation-oriented MRI synthesis. Experimental results demonstrate the superiority of our method both qualitatively and quantitatively in comparison with the state-of-the-art MRI segmentation methods.

Facial pose variation is one of the major factors making face recognition (FR) a challenging task. One popular solution is to convert non-frontal faces to frontal ones on which FR is performed. Rotating faces causes facial pixel value changes. Therefore, existing CNN-based methods learn to synthesize frontal faces in color space. However, this learning problem in a color space is highly non-linear, causing the synthetic frontal faces to lose fine facial textures. In this paper, we take the view that the nonfrontal-frontal pixel changes are essentially caused by geometric transformations (rotation, translation, and so on) in space. Therefore, we aim to learn the nonfrontal-frontal facial conversion in the spatial domain rather than the color domain to ease the learning task. To this end, we propose an appearance-flow-based face frontalization convolutional neural network (A3F-CNN). Specifically, A3F-CNN learns to establish the dense correspondence between the non-frontal and frontal faces. Once the correspondence is built, frontal faces are synthesized by explicitly "moving" pixels from the non-frontal one. In this way, the synthetic frontal faces can preserve fine facial textures. To improve the convergence of training, an appearance-flow-guided learning strategy is proposed. In addition, generative adversarial network loss is applied to achieve a more photorealistic face, and a face mirroring method is introduced to handle the self-occlusion problem. Extensive experiments are conducted on face synthesis and pose invariant FR. Results show that our method can synthesize more photorealistic faces than the existing methods in both the controlled and uncontrolled lighting environments. Moreover, we achieve a very competitive FR performance on the Multi-PIE, LFW and IJB-A databases.