Accurate segmentation of aortic vascular structures is critical for diagnosing and treating cardiovascular diseases. Traditional Transformer-based models have shown promise in this domain by capturing long-range dependencies between vascular features. However, their reliance on fixed-size rectangular patches often influences the integrity of complex vascular structures, leading to suboptimal segmentation accuracy. To address this challenge, we propose the adaptive Morph-Patch Transformer (MPT), a novel architecture specifically designed for aortic vascular segmentation. Specifically, MPT introduces an adaptive patch partitioning strategy that dynamically generates morphology-aware patches aligned with complex vascular structures. This strategy can preserve semantic integrity of complex vascular structures within individual patches. Moreover, a Semantic Clustering Attention (SCA) method is proposed to dynamically aggregate features from various patches with similar semantic characteristics. This method enhances the model's capability to segment vessels of varying sizes, preserving the integrity of vascular structures. Extensive experiments on three open-source datasets (AVT, AortaSeg24 and TBAD) demonstrate that MPT achieves state-of-the-art performance, with improvements in segmenting intricate vascular structures.

Medical image segmentation is a cornerstone of modern clinical diagnostics. While Vision Transformers that leverage shifted window-based self-attention have established new benchmarks in this field, they are often hampered by a critical limitation: their localized attention mechanism struggles to effectively fuse local details with global context. This deficiency is particularly detrimental to challenging tasks such as the segmentation of microtumors and miniature organs, where both finegrained boundary definition and broad contextual understanding are paramount. To address this gap, we propose HBFormer, a novel Hybrid-Bridge Transformer architecture. The 'Hybrid' design of HBFormer synergizes a classic U-shaped encoder-decoder framework with a powerful Swin Transformer backbone for robust hierarchical feature extraction. The core innovation lies in its 'Bridge' mechanism, a sophisticated nexus for multi-scale feature integration. This bridge is architecturally embodied by our novel Multi-Scale Feature Fusion (MFF) decoder. Departing from conventional symmetric designs, the MFF decoder is engineered to fuse multi-scale features from the encoder with global contextual information. It achieves this through a synergistic combination of channel and spatial attention modules, which are constructed from a series of dilated and depth-wise convolutions. These components work in concert to create a powerful feature bridge that explicitly captures long-range dependencies and refines object boundaries with exceptional precision. Comprehensive experiments on challenging medical image segmentation datasets, including multi-organ, liver tumor, and bladder tumor benchmarks, demonstrate that HBFormer achieves state-of-theart results, showcasing its outstanding capabilities in microtumor and miniature organ segmentation. Code and models are available at: https://github.com/lzeeorno/HBFormer.

Accurate 3D vascular segmentation is essential for diagnosing and treating vascular diseases. This task remains challenging due to the complexity of the 3D data and the morphological diversity of blood vessels. In recent years, state space models (SSMs) have received a great attention for its good performance while preserving global receptive field and consuming less computing resources and time. Inspired by this, we propose a model called Swin-VasMamba for 3D vascular segmentation. It consists of a network called CMU-Net and a topologically constrained loss function called dsh loss. We compare our model with several other advanced segmentation models based on CNN, Transformer and Mamba. The results show that Swin-VasMamba achieves a state-of-the-art performance, with the highest Dice coefficient of 0.880, the lowest 95th-percentile of Hausdorff Distance (HD95) of 0.673, and the lowest Average Surface Distance (ASD) of 0.159 on a benchmark dataset.

Three-dimensional blood vessel segmentation is an important and challenging task that faces two main difficulties: (1) blood vessel structures are small, making them hard to capture by the network, and vessel edges are difficult to segment accurately; (2) false positives are prone to occur due to the presence of artifacts and noise. This paper proposes a novel blood vessel segmentation method called MambaVesselNet. This method is based on a state-space model and employs a selective state-space time series modeling strategy to achieve a larger receptive field. To better capture fine vascular structures and accurately segment edges, this paper introduces an edge enhancement module and a feature selection module. In terms of data preprocessing, nnUNet's preprocessing strategy is adopted to ensure spatial consistency of the input data. Evaluation on three standard vascular segmentation benchmarks shows that MambaVesselNet achieves state-of-the-art performance. Specifically, on cardiovascular and liver vessel datasets, the Dice coefficient is improved by 1.38% and 2.69%, respectively. The contributions of this paper include the proposal of a new module for enhancing blood vessel edge features, the development of a feature selection module with long sequence modeling capability, and the adoption of nnUNet's data preprocessing strategy, setting a new benchmark for blood vessel segmentation technology.

Accurate and efficient vertebrae localization and detection in X-ray images are essential for diagnosing and treating spinal diseases. However, most existing methods struggle with the complexity of spine X-ray images, yielding inaccurate results due to insufficient utilization of spinal anatomy information and neglect of individual vertebra characteristics. In this paper, we propose an innovative Vertebrae Localization and Detection Network (VLD-Net) to accurately assist physicians in diagnosing spine-related diseases from X-ray images. Our VLD-Net, for the first time, defines vertebrae localization as a top-bottom sequential decision-making process, employing deep reinforcement learning (DRL) to fully leverage the anatomical information of the spine. Simultaneously, it also prioritizes the distinct characteristics of each vertebra for accurate detection. Specifically, VLD-Net combines three key components: 1) An advanced vertebrae localization module based on DRL is proposed, effectively leveraging anatomical information of the spine. 2) A novel adaptive exploration mechanism is coined to understand the behavior of the DRL agent during training, pinpointing how to effectively achieve the trade-off between exploration and exploitation. 3) An innovative vertebra-focused module is proposed to accurately detect vertebral landmarks, using the attention region of each vertebra as input to enhance focus on the target and reduce interference from surrounding tissue. Extensive experiments on two public spine datasets demonstrate that the VLD-Net outperforms the state-of-the-art methods in accuracy and robustness.

In medical image segmentation, specialized computer vision techniques, notably transformers grounded in attention mechanisms and residual networks employing skip connections, have been instrumental in advancing performance. Nonetheless, previous models often falter when segmenting small, irregularly shaped tumors. To this end, we introduce SMAFormer, an efficient, Transformer-based architecture that fuses multiple attention mechanisms for enhanced segmentation of small tumors and organs. SMAFormer can capture both local and global features for medical image segmentation. The architecture comprises two pivotal components. First, a Synergistic Multi-Attention (SMA) Transformer block is proposed, which has the benefits of Pixel Attention, Channel Attention, and Spatial Attention for feature enrichment. Second, addressing the challenge of information loss incurred during attention mechanism transitions and feature fusion, we design a Feature Fusion Modulator. This module bolsters the integration between the channel and spatial attention by mitigating reshaping-induced information attrition. To evaluate our method, we conduct extensive experiments on various medical image segmentation tasks, including multi-organ, liver tumor, and bladder tumor segmentation, achieving state-of-the-art results. Code and models are available at: https://github.com/lzeeorno/SMAFormer.

Automated segmentation of liver tumors in CT scans is pivotal for diagnosing and treating liver cancer, offering a valuable alternative to labor-intensive manual processes and ensuring the provision of accurate and reliable clinical assessment. However, the inherent variability of liver tumors, coupled with the challenges posed by blurred boundaries in imaging characteristics, presents a substantial obstacle to achieving their precise segmentation. In this paper, we propose a novel dual-branch liver tumor segmentation model, SBCNet, to address these challenges effectively. Specifically, our proposed method introduces a contextual encoding module, which enables a better identification of tumor variability using an advanced multi-scale adaptive kernel. Moreover, a boundary enhancement module is designed for the counterpart branch to enhance the perception of boundaries by incorporating contour learning with the Sobel operator. Finally, we propose a hybrid multi-task loss function, concurrently concerning tumors' scale and boundary features, to foster interaction across different tasks of dual branches, further improving tumor segmentation. Experimental validation on the publicly available LiTS dataset demonstrates the practical efficacy of each module, with SBCNet yielding competitive results compared to other state-of-the-art methods for liver tumor segmentation.

Accurate and automatic delineation of the left atrium (LA) is crucial for computer-aided diagnosis of atrial fibrillation-related diseases. However, effective model training typically requires a large amount of labeled data, which is time-consuming and labor-intensive. In this study, we propose a novel LA delineation framework. The region of LA is first detected using an actor-critic based deep reinforcement learning method with a shape-adaptive detection strategy using only box-level annotations, bypassing the need for voxel-level labeling. With the effectively detected LA, the impacts of class-imbalance and interference from surrounding tissues are significantly reduced. Subsequently, a semi-supervised segmentation scheme is coined to precisely delineate the contour of LA in 3D volume. The scheme integrates two independent networks with distinct structures, enabling implicit consistency regularization, capturing more spatial features, and avoiding the error accumulation present in current mainstream semi-supervised frameworks. Specifically, one network is combined with Transformer to capture latent spatial features, while the other network is based on pure CNN to capture local features. The difference prediction between these two sub-networks is exploited to mutually provide high-quality pseudo-labels and correct the cognitive bias. Experimental results on two public datasets demonstrate that our proposed strategy outperforms several state-of-the-art methods in terms of accuracy and clinical convenience.

Image segmentation achieves significant improvements with deep neural networks at the premise of a large scale of labeled training data, which is laborious to assure in medical image tasks. Recently, semi-supervised learning (SSL) has shown great potential in medical image segmentation. However, the influence of the learning target quality for unlabeled data is usually neglected in these SSL methods. Therefore, this study proposes a novel self-correcting co-training scheme to learn a better target that is more similar to ground-truth labels from collaborative network outputs. Our work has three-fold highlights. First, we advance the learning target generation as a learning task, improving the learning confidence for unannotated data with a self-correcting module. Second, we impose a structure constraint to encourage the shape similarity further between the improved learning target and the collaborative network outputs. Finally, we propose an innovative pixel-wise contrastive learning loss to boost the representation capacity under the guidance of an improved learning target, thus exploring unlabeled data more efficiently with the awareness of semantic context. We have extensively evaluated our method with the state-of-the-art semi-supervised approaches on four public-available datasets, including the ACDC dataset, M&Ms dataset, Pancreas-CT dataset, and Task_07 CT dataset. The experimental results with different labeled-data ratios show our proposed method's superiority over other existing methods, demonstrating its effectiveness in semi-supervised medical image segmentation.

Hard sample selection can effectively improve model convergence by extracting the most representative samples from a training set. However, due to the large capacity of medical images, existing sampling strategies suffer from insufficient exploitation for hard samples or high time cost for sample selection when adopted by 3D patch-based models in the field of multi-organ segmentation. In this paper, we present a novel and effective online hard patch mining (OHPM) algorithm. In our method, an average shape model that can be mapped with all training images is constructed to guide the exploration of hard patches and aggregate feedback from predicted patches. The process of hard mining is formalized as a multi-armed bandit problem and solved with bandit algorithms. With the shape model, OHPM requires negligible time consumption and can intuitively locate difficult anatomical areas during training. The employment of bandit algorithms ensures online and sufficient hard mining. We integrate OHPM with advanced segmentation networks and evaluate them on two datasets containing different anatomical structures. Comparative experiments with other sampling strategies demonstrate the superiority of OHPM in boosting segmentation performance and improving model convergence. The results in each dataset with each network suggest that OHPM significantly outperforms other sampling strategies by nearly 2% average Dice score.

Cerebrovascular segmentation from time-of-flight magnetic resonance angiography (TOF-MRA) data is of great importance in blood supply structure analysis, diagnosis, and treatment of cerebrovascular pathologies. However, complete and accurate segmentation is still a challenge due to the complex image context and vascular morphology. The existing model-driven methods are often difficult to obtain prominent accuracy and robustness. Deep-learning based technology has achieved unimaginable success, but always faces the problem of insufficient labeled data. In this paper, a novel strategy is proposed to automate cerebrovascular segmentation, which integrates model- and data-driven methods. Firstly, the TOF-MRA data are sparsely labeled by three radiologists. Secondly, a semi-supervised mixture probability model is proposed to fit the cerebrovascular intensity distribution precisely, which starts from the sparse annotations and generates massive labeled points. Thirdly, mislabeled points are corrected by a Clean-Mechanism model, to acquire a well-labeled point-set of good quality. Finally, we construct and train a dilated dense convolution network (DD-CNN) by the resultant labeled point-set. The proposed method is validated on 109 clinical TOR-MRA data from a public dataset. Compared with the other state-of-the-art segmentation methods, our method segments cerebrovascular structure with better completeness and sensibility, especially for slender vascularity. The experimental results show that our method reaches an average dice score of 93.20%, which also indicates that the DD-CNN is very competent for cerebrovascular segmentation from TOF-MRA volume.

Guide wire tracking in fluoroscopic images has done a significant task in assisting the physicians during radiology-aided interventions. Many groups have tried to detect the guide wire from the fluoroscopic images based on the image properties. The main challenge is that manual intervention is required during the detection. Other groups try to introduce localizers to track guide wires during intervention, which requires additional hardware equipment, and may intervene with the traditional clinical routines. Machine learning methods are also exploited. Although such methods may provide accurate tracking, they often require large amount of data and training time. In this paper, we propose a GPU-based fast and automatic approach to track guide wires in fluoroscopic sequences. We propose a multi-scale filtering and gradient vector field-based real-time tracking method for guide wire tracking from fluoroscopic images. To improve calculation efficiency and meet real-time application requirement, we propose a GPU-based acceleration scheme, and also a Bayesian filter-like motion tracking method to limit the guide wire tracking to a smaller range to improve calculation efficiency. We test our proposed method on two test data sets of fluoroscopic sequences of 102 frames and 72 frames. We achieve an average guide wire detection rate of 96.7%, a false detection rate of 0.0011% and an error distance measure of 0.83 pixels for the first sequence, and 98.8%, 0.000069% and 0.85 pixels, respectively, for the second sequence. With the proposed acceleration method, we finish calculation for the first sequence in nine seconds, thus, efficiency is enhanced by 100 times with the unaccelerated algorithm.

Active contour model (ACM) plays an important role in computer vision and medical image analysis. The traditional ACMs were employed to extract closed contours of objects. While simultaneous extraction of line- and block-like objects, such as boundary contours, centerlines, as well as their topological relationship, remains open so far. Therefore, a novel ACM named "Ingenious Snake" is proposed to adaptively extract the feature curves. The proposed ingenious snake includes the following steps: 1) In the preprocessing, the line- and block-like structures are classified with k-means clustering, following up with morphological operation and enhancement. The gradient vector flow (GVF) field is then acquired from the resultant ridge feature map. 2) For the automatic initialization, the ridge-points are extracted by using the local phase measurement of GVF field, then the two-category of object ridgelines are obtained fast. 3) Finally, the contour deformation and curves evolvement are implemented with a management strategy. The resultant contours and centerlines well characterize the objects of interest. In the experiments, we compare the existing initialization methods and the adaptive extraction with a series of phantoms and testing images. The visual and quantitative assessments of structure extraction are satisfying in terms of effectiveness and accuracy.

For the segmentation and robust tracking of the cardiac left ventricle (LV) in MRI sequences, an optimized algorithm is presented; it is based on the active contour framework. To use the active contours model (ACM) (Kass, M. et al., Int. J. Comput. Vision, vol.1, p.321-31, 1998) to estimate cardiac motion, a new concept of generalized fuzzy gradient vector flow (GFGVF) is presented and compared with the classical gradient vector flow (GVF) (Chenyang Xu and Prince, J.L., "Gradient Vector Flow Deformable Models", Academic Press, 2000; Chung-Chu Leung and Wufan Chen, Proc. IEEE ICIP Conf., 2003). Then, a modified ACM is proposed for motion tracking, which is based on two new external forces: one is the GFGVF field; the other is the relativity of the optical flow field (OFF) on the predictive contour. For robust tracking of the outline of interest, a set of motion equations is presented to describe two correlative updating steps. Also, given some prior terms and likelihood one, the motion state of each point can be found by the maximum a posteriori probability (MAP).