Long-range Model Research Articles

Retinal artery/vein vessel segmentation and measurement for hypertension based on multi-level edge-guided network

AbstractHypertension is a primary risk factor for the onset of cardiocerebrovascular diseases, leading to increased mortality. In the early stages of hypertension, changes in the diameter of the retinal arteries and veins are closely observed. Therefore, the automatic segmentation of these arteries and veins in retinal fundus images is crucial for hypertension monitoring. However, current deep learning based methods still fail to generate semantically consistent segmentation result, especially for narrow blood vessel segments. Moreover, the pixels near the vessel edge are prone to be missclassified by current methods, struggling to obtain edge-preserving segmentation results. These critical issues severely hinder downstream applications, which require accurate measurement of artery/vein diameter. First, to alleviate the issue of lacking semantic consistency for vessel segments, we propose a long-range spatial dependency modeling module to learn to model the long-range spatial dependency. On this basis, a multi-level edge guided spatial aggregation module is further presented to enhance the ability to accurately classify the edge pixels, generating edge-preserving results. Extensive experimental results on the widely used DRIVE dataset and our constructed dataset (HFC-50) show the superiority of the proposed model over state-of-the-art methods. Our method achieves balanced accuracy of 95.7% on DRIVE dataset, outperforming all state-of-art methods. Finally, in order to directly demonstrate the benefit of more accurate retinal artery/vein vessel segmentation for the measurement of the ratio of artery and vein, we quantitatively evaluate the computed artery/vein ratio for hypertension patients, with the artery/vein segmentation results generated by our proposed method.

Pan-Mamba: Effective pan-sharpening with state space model

Pan-sharpening involves integrating information from low-resolution multi-spectral and high-resolution panchromatic images to generate high-resolution multi-spectral counterparts. While recent advancements in the state space model, particularly the efficient long-range dependency modeling achieved by Mamba, have revolutionized computer vision community, its untapped potential in pan-sharpening motivates our exploration. Our contribution, Pan-Mamba, represents a novel pan-sharpening network that leverages the efficiency of the Mamba model in global information modeling. In Pan-Mamba, we customize two core components: channel swapping Mamba and cross-modal Mamba, strategically designed for efficient cross-modal information exchange and fusion. The former initiates a lightweight cross-modal interaction through the exchange of partial panchromatic and multi-spectral channels, while the latter facilities the information representation capability by exploiting inherent cross-modal relationships. Through extensive experiments across diverse datasets, our proposed approach surpasses state-of-the-art methods, showcasing superior fusion results in pan-sharpening. To the best of our knowledge, this work is the first attempt in exploring the potential of the Mamba model and establishes a new frontier in the pan-sharpening techniques. The source code is available at https://github.com/alexhe101/Pan-Mamba.

ACMamba: A State Space Model-Based Approach for Multi-Weather Degraded Image Restoration

In computer vision, eliminating the effects of adverse weather conditions such as rain, snow, and fog on images is a key research challenge. Existing studies primarily focus on image restoration for single weather types, while methods addressing image restoration under multiple combined weather conditions remain relatively scarce. Furthermore, current mainstream restoration networks, mostly based on Transformer and CNN architectures, struggle to achieve an effective balance between global receptive field and computational efficiency, limiting their performance in practical applications. This study proposes ACMamba, an end-to-end lightweight network based on selective state space models, aimed at achieving image restoration under multiple weather conditions using a unified set of parameters. Specifically, we design a novel Visual State Space Module (VSSM) and a Spatially Aware Feed-Forward Network (SAFN), which organically combine the local feature extraction capabilities of convolutions with the long-range dependency modeling capabilities of selective state space models (SSMs). This combination significantly improves computational efficiency while maintaining a global receptive field, enabling effective application of the Mamba architecture to multi-weather image restoration tasks. Comprehensive experiments demonstrate that our proposed approach significantly outperforms existing methods for both specific and multi-weather tasks across multiple benchmark datasets, showcasing its efficient long-range modeling potential in multi-weather image restoration tasks.

Long-range Ising models: Contours, phase transitions and decaying fields

Inspired by Fröhlich–Spencer and subsequent authors who introduced the notion of contour for long-range systems, we provide a definition of contour and a direct proof for the phase transition for ferromagnetic long-range Ising models on \mathbb{Z}^{d} , d\geq 2 . The argument, which is based on a multiscale analysis, works for the sharp region \alpha>d and improves previous results obtained by Park for \alpha>3d+1 , and by Ginibre, Grossmann and Ruelle for \alpha> d+1 , where \alpha is the power of the coupling constant. The key idea is to avoid a large number of small contours. As an application, we prove the persistence of the phase transition when we add a polynomially decaying magnetic field with power \delta>0 as h^{*}|x|^{-\delta} , where h^{*} >0 . For d<\alpha<d+1 , the phase transition occurs when \delta>\alpha-d , and when h^{*} is small enough over the critical line \delta=\alpha-d . For \alpha \geq d+1 , \delta>1 is enough to prove the phase transition, and for \delta=1 we have to ask h^{*} small. The natural conjecture is that this region is also sharp for the phase transition problem when we have a decaying field.

Influence of Leadership Styles on Performance of Local Government Agencies: Case of Central Equatoria State Local Government in South Sudan

An organization's success depends on successful leadership and good organizational culture. This study investigates the influence of various leadership styles on the performance of local government agencies in Central Equatoria State, Juba, South Sudan. It explores how transformational, transactional, and laissez-faire leadership styles influence the performance of the central Equatoria state local government. The general objective of the study was to examine the influence of various leadership styles under the long-range leadership model on the performance of the central Equatoria state local government. The specific objectives of the study include the following: To assess the influence of transformational leadership style on performance of the central Equatoria state Local Government, to establish the influence of transactional leadership style on performance of the Central Equatoria state Local Government and to examine influence of laisses-faire leadership style on performance of the Central Equatoria state Local Government. The research employed a descriptive design, targeting 791 individuals, including local government officers and leaders from public institutions. A sample of 138 respondents was selected through stratified random sampling, and data was collected via questionnaires, interviews, and document analysis. To ensure validity of the finding, the researcher performed a pilot study on the data collection and analysis instruments in Juba County. This was done to guarantee the instrument's content coverage and authenticity by enhancing problems, the format, and the measuring scales. the researcher ensured protection of the research process to ensure reduction of anomalies, inaccuracy and scoring in the measurements to guarantee reliability of the data obtained. This research was focused on both primary and secondary data. Questionnaires was used as the main tool of data collection supported by both physical and telephone interview. Analysis of reports published on newspapers and internet was as well done. The study utilized the full range leadership model by James MacGregor Burns (1985) and Bernard M. Bass (1985), which encompasses transformational, transactional, and laissez-faire leadership styles. The findings revealed that both transformational and transactional leadership styles are not widely practiced by the CESLG Leaders and had limited impact on performance of the central Equatoria state local government while laissez-faire leadership being the most practiced style, significantly influenced poor performance outcomes of the central Equatoria state local government and it’s one of the reasons for the poor performance of the CESLG institutions. As such the researcher recommend among others, the following to both the CESLG leaders and other stake holders for implementation; The CESLG leaders need to consider recruiting young and well-educated local government officer as the study revealed that majority 48.8% of the local government officials are aged between 51 to 60 years whose performance are average , the local government leaders need to consider initiating executive training and development Programme for local government officers, The CESLG leadership is advised to conduct regular performance review for its institutions and staff. the CESLG leaders and other stakeholders/institutions are encouraged to adopt modern leadership styles such as transformational and transactional in their management styles.

Hyperspectral Image Classification Algorithm for Forest Analysis Based on a Group-Sensitive Selective Perceptual Transformer

Substantial advancements have been achieved in hyperspectral image (HSI) classification through contemporary deep learning techniques. Nevertheless, the incorporation of an excessive number of irrelevant tokens in large-scale remote sensing data results in inefficient long-range modeling. To overcome this hurdle, this study introduces the Group-Sensitive Selective Perception Transformer (GSAT) framework, which builds upon the Vision Transformer (ViT) to enhance HSI classification outcomes. The innovation of the GSAT architecture is primarily evident in several key aspects. Firstly, the GSAT incorporates a Group-Sensitive Pixel Group Mapping (PGM) module, which organizes pixels into distinct groups. This allows the global self-attention mechanism to function within these groupings, effectively capturing local interdependencies within spectral channels. This grouping tactic not only boosts the model’s spatial awareness but also lessens computational complexity, enhancing overall efficiency. Secondly, the GSAT addresses the detrimental effects of superfluous tokens on model efficacy by introducing the Sensitivity Selection Framework (SSF) module. This module selectively identifies the most pertinent tokens for classification purposes, thereby minimizing distractions from extraneous information and bolstering the model’s representational strength. Furthermore, the SSF refines local representation through multi-scale feature selection, enabling the model to more effectively encapsulate feature data across various scales. Additionally, the GSAT architecture adeptly represents both global and local features of HSI data by merging global self-attention with local feature extraction. This integration strategy not only elevates classification precision but also enhances the model’s versatility in navigating complex scenes, particularly in urban mapping scenarios where it significantly outclasses previous deep learning methods. The advent of the GSAT architecture not only rectifies the inefficiencies of traditional deep learning approaches in processing extensive remote sensing imagery but also markededly enhances the performance of HSI classification tasks through the deployment of group-sensitive and selective perception mechanisms. It presents a novel viewpoint within the domain of hyperspectral image classification and is poised to propel further advancements in the field. Empirical testing on six standard HSI datasets confirms the superior performance of the proposed GSAT method in HSI classification, especially within urban mapping contexts, where it exceeds the capabilities of prior deep learning techniques. In essence, the GSAT architecture markedly refines HSI classification by pioneering group-sensitive pixel group mapping and selective perception mechanisms, heralding a significant breakthrough in hyperspectral image processing.

On quasi-locality and decay of correlations for long-range models of open quantum spin systems

Abstract We consider models of open quantum spin systems with irreversible dynamics and show
that general quasi-locality results for long-range models, e.g. as proven for the Heisenberg dynamics
associated to quantum systems in [1], naturally extend to this setting. 
Given these bounds, we provide two applications. First, we use these results to obtain 
estimates on a strictly local approximation of these finite-volume, irreversible dynamics.
Next, we show how these bounds can be used to estimate correlation decay in various states.

How Fast do Rumours Spread?

We study a rumour propagation model along the lines of Lebensztayn and Rodriguez (Stat Probab Lett 78(14):2130–2136, 2008) as a long-range percolation model on Z\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\mathbb {Z}$$\\end{document}. We begin by showing a sharp phase transition-type behaviour in the sense of exponential decay of the survival time of the rumour cluster in the sub-critical phase. In the super-critical phase, under the assumption that radius of influence r.v. has 2+ϵ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$2+\\epsilon $$\\end{document} moment finite (for some ϵ>0\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\epsilon >0$$\\end{document}), we show that the rightmost vertex in the rumour cluster has a deterministic speed in the sense that after appropriate scaling, the location of the rightmost vertex converges a.s. to a deterministic positive constant. Under the assumption that radius of influence r.v. has 4+ϵ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$4+\\epsilon $$\\end{document} moment finite, we obtain a central limit theorem for appropriately scaled and centered rightmost vertex. Later, we introduce a rumour propagation model with reactivation. For this section, we work with a family of exponentially decaying i.i.d. radius of influence r.v.’s, and we obtain the speed result for the scaled rightmost position of the rumour cluster. Each of these results is novel, in the sense that such properties have never been established before in the context of the rumour propagation model on Z\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\mathbb {Z}$$\\end{document}, to the best of our knowledge.

Non-stationary mechanical sound source separation: An all-neural beamforming network driven by time–frequency convolution and self-attention

To tackle the challenging task of extracting target signal features amidst significant interference from complex, time-varying non-stationary noise in industrial scenarios, we propose a novel time-domain all-neural beamforming network tailored for non-stationary mechanical sound source separation, named TFANBNet. Leveraging time–frequency convolution and self-attention, TFANBNet promised robust performance in challenging acoustic environments. First, the proposed time–frequency convolution module employs parameterized complex-valued convolutional layers to simulate time–frequency transformations, thereby implicitly extracting time–frequency features. Then, the proposed adaptive attention transformer network enhances local attention interaction by integrating intermediate features, effectively enhancing long-range context modeling. Third, the beamforming module utilizes deep neural networks for beamforming weight estimation, replacing conventional noise covariance matrix computations, which have inherent performance limitations. Finally, the experimental results on the multi-channel spatial reverberation dataset synthesized from the MIMII dataset show that the proposed TFANBNet yields superior separation performance and generalization capabilities compared to the considered competitive methods.

Probing electron localization and delocalization in the selective long-range tight-binding model

Probing electron localization and delocalization in the selective long-range tight-binding model

High-resolution acoustically informed maps of sound speed.

As oceanographic models advance in complexity, accuracy, and resolution, in situ measurements must provide spatiotemporal information with sufficient resolution to inform and validate those models. In this study, water masses at the New England shelf break were mapped using scientific echosounders combined with water column property measurements from a single conductivity, temperature, and depth (CTD) profile. The acoustically-inferred map of sound speed was compared with a sound speed cross section based on two-dimensional interpolation of multiple CTD profiles. Long-range acoustic propagation models were then parameterized by the sound speed profiles estimated by the two methods and differences were compared.

Retrieval-Pretrained Transformer: Long-range Language Modeling with Self-retrieval

Abstract Retrieval-augmented language models (LMs) have received much attention recently. However, typically the retriever is not trained jointly as a native component of the LM, but added post-hoc to an already-pretrained LM, which limits the ability of the LM and the retriever to adapt to one another. In this work, we propose the Retrieval-Pretrained Transformer (RPT), an architecture and training procedure for jointly training a retrieval-augmented LM from scratch and applying it to the task of modeling long texts. Given a recently generated text chunk in a long document, the LM computes query representations, which are then used to retrieve earlier chunks in the document, located potentially tens of thousands of tokens before. Information from retrieved chunks is fused into the LM representations to predict the next target chunk. We train the retriever component with a semantic objective, where the goal is to retrieve chunks that increase the probability of the next chunk, according to a reference LM. We evaluate RPT on four long-range language modeling tasks, spanning books, code, and mathematical writing, and demonstrate that RPT improves retrieval quality and subsequently perplexity across the board compared to strong baselines.

Thin Cloud Removal Generative Adversarial Network Based on Sparse Transformer in Remote Sensing Images

Thin clouds in Remote Sensing (RS) imagery can negatively impact subsequent applications. Current Deep Learning (DL) approaches often prioritize information recovery in cloud-covered areas but may not adequately preserve information in cloud-free regions, leading to color distortion, detail loss, and visual artifacts. This study proposes a Sparse Transformer-based Generative Adversarial Network (SpT-GAN) to solve these problems. First, a global enhancement feature extraction module is added to the generator’s top layer to enhance the model’s ability to preserve ground feature information in cloud-free areas. Then, the processed feature map is reconstructed using the sparse transformer-based encoder and decoder with an adaptive threshold filtering mechanism to ensure sparsity. This mechanism enables that the model preserves robust long-range modeling capabilities while disregarding irrelevant details. In addition, inverted residual Fourier transformation blocks are added at each level of the structure to filter redundant information and enhance the quality of the generated cloud-free images. Finally, a composite loss function is created to minimize error in the generated images, resulting in improved resolution and color fidelity. SpT-GAN achieves outstanding results in removing clouds both quantitatively and visually, with Structural Similarity Index (SSIM) values of 98.06% and 92.19% and Peak Signal-to-Noise Ratio (PSNR) values of 36.19 dB and 30.53 dB on the RICE1 and T-Cloud datasets, respectively. On the T-Cloud dataset, especially with more complex cloud components, the superior ability of SpT-GAN to restore ground details is more evident.

Iterative Mamba Diffusion Change-Detection Model for Remote Sensing

In the field of remote sensing (RS), change detection (CD) methods are critical for analyzing the quality of images shot over various geographical areas, particularly for high-resolution images. However, there are some shortcomings of the widely used Convolutional Neural Networks (CNNs) and Transformers-based CD methods. The former is limited by its insufficient long-range modeling capabilities, while the latter is hampered by its computational complexity. Additionally, the commonly used information-fusion methods for pre- and post-change images often lead to information loss or redundancy, resulting in inaccurate edge detection. To address these issues, we propose an Iterative Mamba Diffusion Change Detection (IMDCD) approach to iteratively integrate various pieces of information and efficiently produce fine-grained CD maps. Specifically, the Swin-Mamba-Encoder (SME) within Mamba-CD (MCD) is employed as a semantic feature extractor, capable of modeling long-range relationships with linear computability. Moreover, we introduce the Variable State Space CD (VSS-CD) module, which extracts abundant CD features by training the matrix parameters within the designed State Space Change Detection (SS-CD). The computed high-dimensional CD feature is integrated into the noise predictor using a novel Global Hybrid Attention Transformer (GHAT) while low-dimensional CD features are utilized to calibrate prior CD results at each iterative step, progressively refining the generated outcomes. IMDCD exhibits a high performance across multiple datasets such as the CDD, WHU, LEVIR, and OSCD, marking a significant advancement in the methodologies within the CD field of RS. The code for this work is available on GitHub.

DeConformer-SENet: An efficient deformable conformer speech enhancement network

The Conformer model has demonstrated superior performance in speech enhancement by combining the long-range relationship modeling capability of self-attention with the local information processing ability of convolutional neural networks (CNNs). However, existing Conformer-based speech enhancement models struggle to balance performance and model complexity. In this work, we propose, DeConformer-SENet, an end-to-end time-domain deformable Conformer speech enhancement model, with modifications to both the self-attention and CNN components. Firstly, we introduce the time-frequency-channel self-attention (TFC-SA) module, which compresses information from each dimension of the input features into a one-dimensional vector. By calculating the energy distribution, this module models long-range relationships across three dimensions, reducing computational complexity while maintaining performance. Additionally, we replace standard convolutions with deformable convolutions, aiming to expand the receptive field of the CNN and accurately model local features. We validate our proposed DeConformer-SENet on the WSJ0-SI84 + DNS Challenge dataset. Experimental results demonstrate that DeConformer-SENet outperforms existing Conformer and Transformer models in terms of ESTOI and PESQ metrics, while also being more computationally efficient. Furthermore, ablation studies confirm that DeConformer-SENet improvements enhance the performance of conventional Conformer and reduce model complexity without compromising the overall effectiveness.

Learning long sequences in spiking neural networks

Spiking neural networks (SNNs) take inspiration from the brain to enable energy-efficient computations. Since the advent of Transformers, SNNs have struggled to compete with artificial networks on modern sequential tasks, as they inherit limitations from recurrent neural networks (RNNs), with the added challenge of training with non-differentiable binary spiking activations. However, a recent renewed interest in efficient alternatives to Transformers has given rise to state-of-the-art recurrent architectures named state space models (SSMs). This work systematically investigates, for the first time, the intersection of state-of-the-art SSMs with SNNs for long-range sequence modelling. Results suggest that SSM-based SNNs can outperform the Transformer on all tasks of a well-established long-range sequence modelling benchmark. It is also shown that SSM-based SNNs can outperform current state-of-the-art SNNs with fewer parameters on sequential image classification. Finally, a novel feature mixing layer is introduced, improving SNN accuracy while challenging assumptions about the role of binary activations in SNNs. This work paves the way for deploying powerful SSM-based architectures, such as large language models, to neuromorphic hardware for energy-efficient long-range sequence modelling.

Ordering kinetics with long-range interactions: interpolating between voter and Ising models

Abstract We study the ordering kinetics of a generalization of the voter model with long-range interactions, the p-voter model, in one dimension. It is defined in terms of Boolean variables S i , agents or spins, located on sites i of a lattice, each of which takes in an elementary move the state of the majority of p other agents at distances r chosen with probability P ( r ) ∝ r − α . For p = 2 the model can be exactly mapped onto the case with p = 1, which amounts to the voter model with long-range interactions decaying algebraically. For 3 ⩽ p < ∞ , instead, the dynamics falls into the universality class of the one-dimensional Ising model with long-ranged coupling constant J ( r ) = P ( r ) quenched to small finite temperatures. In the limit p → ∞ , a crossover to the (different) behavior of the long-range Ising model quenched to zero temperature is observed. Since for p > 3 a closed set of differential equations cannot be found, we employed numerical simulations to address this case.

Lightweight image super-resolution with sliding Proxy Attention Network

Recently, image super-resolution (SR) models using window-based Transformers have demonstrated superior performance compared to SR models based on convolutional neural networks. Nevertheless, Transformer-based SR models often entail high computational demands. This is due to the adoption of shifted window self-attention following the window self-attention layer to model long-range relationships, resulting in additional computational overhead. Moreover, extracting local image features only with the self-attention mechanism is insufficient to reconstruct rich high-frequency image content. To overcome these challenges, we propose the Sliding Proxy Attention Network (SPAN), capable of recovering high-quality High-Resolution (HR) images from Low-Resolution (LR) inputs with substantially fewer model parameters and computational operations. The primary innovation of SPAN lies in the Sliding Proxy Transformer Block (SPTB), integrating the local detail sensitivity of convolution with the long-range dependency modeling of self-attention mechanism. Key components within SPTB include the Enhanced Local Feature Extraction Block (ELFEB) and the Sliding Proxy Attention Block (SPAB). ELFEB is designed to enhance the local receptive field with lightweight parameters for high-frequency details compensation. SPAB optimizes computational efficiency by implementing intra-window and cross-window attention in a single operation through leveraging window overlap. Experimental results demonstrate that SPAN can produce high-quality SR images while effectively managing computational complexity. The code is publicly available at: https://github.com/zononhzy/SPAN.

Muon g−2 , long-range muon spin force, and neutrino oscillations

Recent studies have proposed using a geocentric muon spin force to account for the (g−2)μ anomaly, with the long-range force mediator being a light axionlike particle. The mediator exhibits a CP-violating scalar coupling to nucleons and a normal derivative coupling to muons. Due to the weak symmetry, this axion inevitably couples to neutrinos, providing potential impact on neutrino oscillations. By utilizing neutrino data from BOREXINO, IceCube DeepCore, Super-Kamiokande, and SNO, we have identified that both atmospheric and solar neutrino data can impose stringent constraints on the long-range muon spin force model and the (g−2)μ parameter space. With optimized data analysis techniques and the potential from future experiments, such as JUNO, Hyper-Kamiokande, SNO+, and IceCube PINGU, there exists a promising opportunity to achieve even greater sensitivities. Indeed, neutrino oscillations offer a robust and distinctive cross-check for the model, offering stringent constraints on the (g−2)μ parameter space. Published by the American Physical Society 2024

A mixed Mamba U-net for prostate segmentation in MR images

The diagnosis of early prostate cancer depends on the accurate segmentation of prostate regions in magnetic resonance imaging (MRI). However, this segmentation task is challenging due to the particularities of prostate MR images themselves and the limitations of existing methods. To address these issues, we propose a U-shaped encoder-decoder network MM-UNet based on Mamba and CNN for prostate segmentation in MR images. Specifically, we first proposed an adaptive feature fusion module based on channel attention guidance to achieve effective fusion between adjacent hierarchical features and suppress the interference of background noise. Secondly, we propose a global context-aware module based on Mamba, which has strong long-range modeling capabilities and linear complexity, to capture global context information in images. Finally, we propose a multi-scale anisotropic convolution module based on the principle of parallel multi-scale anisotropic convolution blocks and 3D convolution decomposition. Experimental results on two public prostate MR image segmentation datasets demonstrate that the proposed method outperforms competing models in terms of prostate segmentation performance and achieves state-of-the-art performance. In future work, we intend to enhance the model's robustness and extend its applicability to additional medical image segmentation tasks.