Entropy-guided semi-supervised framework for robust chest X-ray segmentation using dynamic competition and patch-wise contrastive learning

In this study, we propose a novel point cloud based 3D registration and segmentation framework using reinforcement learning. An artificial agent, implemented as a distinct actor based on value networks, is trained to predict the optimal piece-wise linear transformation of a point cloud for the joint tasks of registration and segmentation. The actor network estimates a set of plausible actions and the value network aims to select the optimal action for the current observation. Point-wise features that comprise spatial positions (and surface normal vectors in the case of structured meshes), and their corresponding image features, are used to encode the observation and represent the underlying 3D volume. The actor and value networks are applied iteratively to estimate a sequence of transformations that enable accurate delineation of object boundaries. The proposed approach was extensively evaluated in both segmentation and registration tasks using a variety of challenging clinical datasets. Our method has fewer trainable parameters and lower computational complexity compared to the 3D U-Net, and it is independent of the volume resolution. We show that the proposed method is applicable to mono- and multi-modal segmentation tasks, achieving significant improvements over the state-of-the-art for the latter. The flexibility of the proposed framework is further demonstrated for a multi-modal registration application. As we learn to predict actions rather than a target, the proposed method is more robust compared to the 3D U-Net when dealing with previously unseen datasets, acquired using different protocols or modalities. As a result, the proposed method provides a promising multi-purpose segmentation and registration framework, particular in the context of image-guided interventions.

Nowadays under human observations iris recognition is playing an essential role, it observed that there are some points are very essential for this method like accuracy, efficiency & processing time. There are several problems has been recorded for this kind of technique. Under previous researches, many concepts have been reported for iris-based segmentation, classification & feature extraction techniques. This research dealt with implementing a novel framework of robust iris segmentation utilizing computer vision technique, FFNN classifier & hybrid feature extraction technique. For the non-iris images method of the novel iris, segmentation has been proposed in this paper. The technique of novel iris segmentation based on 2 methods regards pupil segmentation, after that fusion of enhancing & shrinking visible contour has been framed for segmentation of iris through collaborating novel burden forcing for visible contour model. Moreover, then UN wrapped iris segmentation performed well with the technique of normalization non-circular iris. Whatever, for better iris segmentation research, utilized a proposed method of feature extraction under Discrete Wavelet Transform, letter utilized geometric & texture features of the segmented image. Achieved features have collaborated through a vector of a hybrid feature. For the reason of recognition, we are utilizing a vision algorithm with an FFNN classifier. This paper focused on the concepts related to a robust iris recognition framework using a vision algorithm. Daugman demodulates the yield of the Gabor channels to pack the information. It separates the stage data into four levels and for every conceivable quadrant in-plane to get a minimal 256-byte layout, which takes into consideration proficient capacity and correlation of irises. Experimental outputs have been described as the proposed robust iris segmentation method adopted greater accuracy on the challenging segmentation algorithm thousand databases.

Histopathological diagnosis remains the gold standard for confirming gastric tubular adenocarcinoma. As a central task in computer-aided diagnosis, pathological image segmentation plays a crucial role in enhancing both diagnostic efficiency and accuracy. Microscopic hyperspectral images (MHSIs) have demonstrated substantial potential by providing distinctive spectral signatures, while conventional RGB images offer richer spatial textures and fine-grained structural details. The effective fusion of these complementary modalities can significantly boost segmentation performance. However, current deep learningbased multimodal segmentation methods often fail to fully exploit the unique attributes of each modality. These approaches mostly rely on complex fusion architectures with limited interpretability, ultimately leading to suboptimal outcomes. To address these challenges, we propose a novel handcrafted feature-guided multimodal fusion segmentation framework for gastric pathology segmentation. Our approach leverages statistically meaningful handcrafted spatial features extracted from RGB images to guide the deep network in learning more modality-sensitive and discriminative representations. Considering the structural discrepancies between 2D RGB and 3D MHSI, we propose three feature fusion strategies that enhance cross-modal complementarity and support efficient and robust segmentation. Experiments on a multimodal gastric pathology dataset demonstrate that our method effectively segments tumor, stroma, and space regions. Compared to TransUNet, it achieves a <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 0. 1 \%}$</tex> increase in IoU, and a 7.4 % reduction in Hausdorff Distance. Ablation studies further confirm that handcrafted feature guidance consistently enhances segmentation performance across all three fusion strategies.

This paper presents a new Bayesian framework for layered motion segmentation, dividing the frames of an image sequence into foreground and background layers by tracking edges. The first frame in the sequence is segmented into regions using image edges, which are tracked to estimate two affine motions. The probability of the edges fitting each motion is calculated using 1st order statistics along the edge. The most likely region labelling is then resolved using these probabilities, together with a Markov Random Field prior. As part of this process one of the motions is also identified as the foreground motion. Good results are obtained using only two frames for segmentation. However, it is also demonstrated that over multiple frames the probabilities may be accumulated to provide an even more accurate and robust segmentation. The final region labelling can be used, together with the two motion models, to produce a good segmentation of an extended sequence.

Deformable models have had great successes over the past 20 years in medical applications. We have recently developed new classes of deformable models which we term hybrid deformable models to automate the model initialization process and make improvements in segmentation and registration. In this paper we present several hybrid deformable methods we have been developing for segmentation and registration. These methods include Metamorphs, a novel shape and texture integration deformable model framework and the integration of deformable models with graphical models and learning methods. We first present a framework for the robust segmentation and tracking of the heart from tagged MRI images and second applications involving brain tumor segmentation as well as brain and cardiac shape registration.

An automated segmentation framework for nasal computational fluid dynamics analysis in computed tomography

Robust organ segmentation is a prerequisite for computer-aided diagnosis, quantitative imaging analysis, pathology detection, and surgical assistance. For organs with high anatomical variability (e.g., the pancreas), previous segmentation approaches report low accuracies, compared with well-studied organs, such as the liver or heart. We present an automated bottom-up approach for pancreas segmentation in abdominal computed tomography (CT) scans. The method generates a hierarchical cascade of information propagation by classifying image patches at different resolutions and cascading (segments) superpixels. The system contains four steps: 1) decomposition of CT slice images into a set of disjoint boundary-preserving superpixels; 2) computation of pancreas class probability maps via dense patch labeling; 3) superpixel classification by pooling both intensity and probability features to form empirical statistics in cascaded random forest frameworks; and 4) simple connectivity based post-processing. Dense image patch labeling is conducted using two methods: efficient random forest classification on image histogram, location and texture features; and more expensive (but more accurate) deep convolutional neural network classification, on larger image windows (i.e., with more spatial contexts). Over-segmented 2-D CT slices by the simple linear iterative clustering approach are adopted through model/parameter calibration and labeled at the superpixel level for positive (pancreas) or negative (non-pancreas or background) classes. The proposed method is evaluated on a data set of 80 manually segmented CT volumes, using six-fold cross-validation. Its performance equals or surpasses other state-of-the-art methods (evaluated by "leave-one-patient-out"), with a dice coefficient of 70.7% and Jaccard index of 57.9%. In addition, the computational efficiency has improved significantly, requiring a mere 6 ~ 8 min per testing case, versus ≥ 10 h for other methods. The segmentation framework using deep patch labeling confidences is also more numerically stable, as reflected in the smaller performance metric standard deviations. Finally, we implement a multi-atlas label fusion (MALF) approach for pancreas segmentation using the same data set. Under six-fold cross-validation, our bottom-up segmentation method significantly outperforms its MALF counterpart: 70.7±13.0% versus 52.51±20.84% in dice coefficients.

Noisy labels collected with limited annotation cost prevent medical image segmentation algorithms from learning precise semantic correlations. Previous segmentation arts of learning with noisy labels merely perform a pixel-wise manner to preserve semantics, such as pixel-wise label correction, but neglect the pair-wise manner. In fact, we observe that the pair-wise manner capturing affinity relations between pixels can greatly reduce the label noise rate. Motivated by this observation, we present a novel perspective for noisy mitigation by incorporating both pixel-wise and pair-wise manners, where supervisions are derived from noisy class and affinity labels, respectively. Unifying the pixel-wise and pair-wise manners, we propose a robust Joint Class-Affinity Segmentation (JCAS) framework to combat label noise issues in medical image segmentation. Considering the affinity in pair-wise manner incorporates contextual dependencies, a differentiated affinity reasoning (DAR) module is devised to rectify the pixel-wise segmentation prediction by reasoning about intra-class and inter-class affinity relations. To further enhance the noise resistance, a class-affinity loss correction (CALC) strategy is designed to correct supervision signals via the modeled noise label distributions in class and affinity labels. Meanwhile, CALC strategy interacts the pixel-wise and pair-wise manners through the theoretically derived consistency regularization. Extensive experiments under both synthetic and real-world noisy labels corroborate the efficacy of the proposed JCAS framework with a minimum gap towards the upper bound performance. The source code is available at https://github.com/CityU-AIM-Group/JCAS.KeywordsClass and affinityLoss correctionNoisy label

Automated segmentation of epidermis in high-frequency ultrasound of pathological skin using a cascade of DeepLab v3+ networks and fuzzy connectedness.

Mitochondria segmentation from electron microscopy (EM) images plays a crucial role in biological and medical research. However, models trained on source domains often suffer from performance degradation when applied to target domains due to domain shift. Unsupervised domain adaptation (UDA) methods have been proposed to address this issue, but they often overlook the reliability of pseudo-labels and the effectiveness of supervision signals. In this paper, we propose R4MITO, a novel UDA framework for robust mitochondria segmentation. First, we introduce Reliable Prototype Pseudo-labels to mitigate the inconsistency of class-level features between across domains by leveraging source prototypes to model target prototypes. Second, we devise Correlation-wise Consistency Regularization to exploit inter-pixel correlations, aligning agent-level correlations under various perturbations. Third, we propose Rank-aware Relationship Consistency Regularization to fully utilize the rich information encoded in inter-agent relationships by imposing rank-aware constraints on agent-ranking probability distributions. Extensive experiments on multiple EM datasets demonstrate the superiority of our R4MITO over existing state-of-the-art UDA methods for mitochondria segmentation.

A multi-scale geometric feature-adaptive density-aware framework for robust sub-conductor segmentation in high-voltage transmission corridors

We propose an optimization framework for interactive image segmentation (IIS) that operates in bilateral space to achieve robust object extraction and instant visual feedback. More specifically, we first resample an input image using a regular bilateral grid with a resolution that is typically coarser than the input image to reduce the complexity of subsequent IIS tasks. We then design a Markov random field energy on the vertices of the bilateral grid that can be solved efficiently using a standard graph cut label assignment. To achieve this, we introduce reliable color models to distinguish the foreground and background despite the presence of extremely difficult cases and a higher-order potential to encourage spatial consistency in segmentation. We conduct comprehensive experiments on three standard interactive segmentation datasets, MSRA 10K, IIS, and PASCAL VOC 2012 segmentation validation set. The results show that the proposed method achieves competitive performance compared with state-of-the-art methods while making the current system efficient in terms of speed.

Atlas-based liver segmentation and hepatic fat-fraction assessment for clinical trials

Enhanced particle-filtering framework for vessel segmentation and tracking

Referring Video Object Segmentation (RVOS) aims to segment objects in videos based on natural language descriptions, which requires accurate spatial localization and temporal consistency. In this work, we propose a novel framework for robust video object segmentation by modifying the textual backbone of Grounding DINO in combination with an enhanced SAM2 model integrated with Iterative Kalman Filtering (IKF). Our approach leverages the open-ended lexical detection capabilities of Grounding DINO to recognize target objects described by natural language, while IKF-enhanced SAM2 ensures temporal consistency and accurate segmentation across video frames. With extensive experiments on Mevis and other datasets, our approach performs well in challenging scenarios with occlusion, fast motion, and complex scene changes while maintaining performance.