The inherently low signal-to-noise ratio (SNR) in diffusion-weighted (DW) imaging fundamentally impedes precise tissue microstructure characterization, rendering effective noise suppression a persistent challenge. Existing denoising methods frequently suffer from over-smoothing or distortion of microstructure information when handling spatially correlated or severe noise. To address these limitations, we propose UP2-MAE fusion model, a self-supervised DWI denoising method based on Uncertainty-Propelled Physics and Masked Auto-Encoder (MAE) fusion. This framework integrates two complementary branches: one leverages MAE to suppress noise through local context modeling, while the other constructs uncorrelated noisy pairs using diffusion tensor imaging (DTI) physics and denoises them via a Noise2Noise approach, which can preserve texture details by exploiting directional relationships across diffusion encoding directions. To fully integrate the strengths of both branches, an uncertainty-propelled fusion strategy based on maximum likelihood estimation is proposed to derive the final denoised output. In addition, to further promote the performance, uncertainty-guided reconstruction and consistency loss are presented. Evaluations against state-of-the-art denoising methods on both simulated and acquired DW datasets confirm the efficacy of our approach.

Multi-contrast magnetic resonance imaging (MRI) super-resolution intends to reconstruct high-resolution (HR) images from low-resolution (LR) scans by leveraging structural information present in HR reference images acquired with different contrasts. This technique enhances anatomical detail and soft tissue differentiation, which is vital for early diagnosis and clinical decision-making. However, inherent contrasts disparities between modalities pose fundamental challenges in effectively utilizing reference image textures to guide target image reconstruction, often resulting in suboptimal feature integration. To address this issue, we propose a dual-prompt expert network based on a convolutional dictionary feature decoupling (CD-DPE) strategy for multi-contrast MRI super-resolution. Specifically, we introduce an iterative convolutional dictionary feature decoupling module (CD-FDM) to separate features into cross-contrast and intra-contrast components, thereby reducing redundancy and interference. To fully integrate these features, a novel dual-prompt feature fusion expert module (DP-FFEM) is proposed. This module uses a frequency prompt to guide the selection of relevant reference features for incorporation into the target image, while an adaptive routing prompt determines the optimal method for fusing reference and target features to enhance reconstruction quality. Extensive experiments on public multi-contrast MRI datasets demonstrate that CD-DPE outperforms state-of-the-art methods in reconstructing fine details. Additionally, experiments on unseen datasets demonstrated that CD-DPE exhibits strong generalization capabilities.

Although the widespread use of Large Language Models (LLMs) brings convenience, it also raises concerns about the credibility of academic research and scholarly processes. To better understand the extent and characteristics of LLM use in scholarly writing and peer review, the penetration of LLMs across academic workflows is evaluated from multiple perspectives and dimensions, providing compelling evidence of their growing influence. A framework consisting of two components is proposed: ScholarLens , a curated dataset of human-written and LLM-generated content across scholarly writing and peer review for multi-perspective evaluation, and LLMetrica , a tool for assessing LLM penetration using rule-based metrics and model-based detectors for multi-dimensional evaluation. The effectiveness of LLMetrica is demonstrated through experiments, revealing the increasing role of LLMs in scholarly processes. These findings emphasize the need for transparency, accountability, and ethical practices in the use of LLMs to maintain academic credibility.

Fish feeding intensity assessment (FFIA) aims to evaluate fish appetite changes during feeding, which is crucial in industrial aquaculture applications. Existing FFIA methods are limited by their robustness to noise, computational complexity, and the lack of public datasets for developing the models. To address these issues, we first introduce AV-FFIA, a new dataset containing 27,000 labeled audio and video clips that capture different levels of fish feeding intensity. Then, we introduce multi-modal approaches for FFIA by leveraging the models pre-trained on individual modalities and fused with data fusion methods. We perform benchmark studies of these methods on AV-FFIA, and demonstrate the advantages of the multi-modal approach over the single-modality based approach, especially in noisy environments. However, compared to the methods developed for individual modalities, the multimodal approaches may involve higher computational costs due to the need for independent encoders for each modality. To overcome this issue, we further present a novel unified mixed-modality based method for FFIA, termed as U-FFIA. U-FFIA is a single model capable of processing audio, visual, or audio-visual modalities, by leveraging modality dropout during training and knowledge distillation using the models pre-trained with data from single modality. We demonstrate that U-FFIA can achieve performance better than or on par with the state-of-the-art modality-specific FFIA models, with significantly lower computational overhead, enabling robust and efficient FFIA for improved aquaculture management. To encourage further research, we have released the AV-FFIA dataset, the pre-trained model and codes athttps://github.com/FishMaster93/U-FFIA. Note to Practitioners—Feeding is one of the most important costs in aquaculture. However, current feeding machines usually operate with fixed thresholds or human experiences, lacking the ability to automatically adjust to fish feeding intensity. FFIA can evaluate the intensity changes in fish appetite during the feeding process and optimize the control strategies of the feeding machine to avoid inadequate feeding or overfeeding, thereby reducing the feeding cost and improving the well-being of fish in industrial aquaculture. The existing methods have mainly exploited single-modality data, and have a high sensitivity to input noise. Using video and audio offers improved chances to address the challenges brought by various environments. However, compared with processing data from single modalities, using multiple modalities simultaneously often involves increased computational resources, including memory, processing power, and storage. This can impact system performance and scalability. To address these issues, we focus on the efficient unified model, which is capable of processing both multimodal and single-modal input. Our proposed model achieved state-of-the-art (SOTA) performance in FFIA with high computational efficiency.

High-quality 3D fetal brain MRI reconstruction from motion-corrupted 2D slices is crucial for precise clinical diagnosis and advancing our understanding of fetal brain development. This necessitates reliable slice-to-volume registration (SVR) for motion correction and super-resolution reconstruction (SRR) techniques. Traditional approaches have their limitations, but deep learning (DL) offers the potential in enhancing SVR and SRR. However, most of DL methods require large-scale external 3D high-resolution (HR) training datasets, which is challenging in clinical fetal MRI. To address this issue, we propose an unsupervised iterative joint SVR and SRR DL framework for 3D isotropic HR volume reconstruction. Specifically, our method conceptualizes SVR as a function that maps a 2D slice and a 3D target volume to a rigid transformation matrix, aligning the slice to the underlying location within the target volume. This function is parameterized by a convolutional neural network, which is trained by minimizing the difference between the volume slicing at the predicted position and the actual input slice. For SRR, a decoding network embedded within a deep image prior framework, coupled with a comprehensive image degradation model, is used to produce the HR volume. The deep image prior framework offers a local consistency prior to guide the reconstruction of HR volumes. By performing a forward degradation model, the HR volume is optimized by minimizing the loss between the predicted slices and the acquired slices. Experiments on both large-magnitude motion-corrupted simulation data and clinical data have shown that our proposed method outperforms current state-of-the-art fetal brain reconstruction methods.

Low signal to noise ratio (SNR) remains one of the limitations of diffusion weighted (DW) imaging. How to suppress the influence of noise on the subsequent analysis about the tissue microstructure is still challenging. This work proposed a novel self-supervised learning model, Replace2Self, to effectively reduce spatial correlated noise in DW images. Specifically, a voxel replacement strategy based on similar block matching in Q-space was proposed to destroy the correlations of noise in DW image along one diffusion gradient direction. To alleviate the signal gap caused by the voxel replacement, an image mixing strategy based on complementary mask was designed to generate two different noisy DW images. After that, these two noisy DW images were taken as input, and the non-correlated noisy DW image after voxel replacement was taken as learning target, a denoising network was trained for denoising. To promote the denoising performance, a complementary mask mixing consistency loss and an inverse replacement regularization loss were also proposed. Through the comparisons against several existing DW image denoising methods on extensive simulation data with different noise distributions, noise levels and b-values, as well as the acquisition datasets and the ablation experiments, we verified the effectiveness of the proposed method. Regardless of the noise distribution and noise level, the proposed method achieved the highest PSNR, which was at least 1.9% higher than the suboptimal method when the noise level reaches 10%. Furthermore, our method has superior generalization ability due to the use of the proposed strategies.

Considering the tumor aggressive nature and the significant changes in anatomical structure, aligning the preoperative and follow up scans of glioma patients remains a challenge due to the presence of regions with absent correspondence. To address this challenge, this work proposed a novel bidirectional unsupervised deformable image registration framework for image pairs with missing correspondence based on an auxiliary-image-aided intensity-consistency constraint (ICC) strategy. Specifically, for any fixed and moving image pairs, we introduced an auxiliary image and warped it directly to fixed/moving image or warped it twice through a transition of moving/fixed image. By comparing the difference between these warped images, the weighting maps to identify and exclude regions with absent correspondence between fixed and moving image pairs can be generated. To verify the effectiveness of the proposed framework, we combined it with several deep learning-based registration models and tested it on BraTS-Reg challenge dataset, the results demonstrated that the proposed ICC strategy can improve the registration performance for all the models, with the improvement of average target registration error (TRE) and success rate (SR) being up to 44.9% and 66.7%, respectively. Comparing against the best existing forward-backward consistency strategy for dealing with missing correspondence registration, our auxiliary-image-aided ICC strategy can also decrease average TRE by 2.9%, demonstrating the superiority of the proposed framework. The present work is not limited to the glioma images, it can be used to address the registration problems for any image pairs with absent correspondence or inconsistent intensity.

To promote the generalization ability of breast tumor segmentation models, as well as to improve the segmentation performance for breast tumors with smaller size, low-contrast and irregular shape, we propose a progressive dual priori network (PDPNet) to segment breast tumors from dynamic enhanced magnetic resonance images (DCE-MRI) acquired at different centers. The PDPNet first cropped tumor regions with a coarse-segmentation based localization module, then the breast tumor mask was progressively refined by using the weak semantic priori and cross-scale correlation prior knowledge. To validate the effectiveness of PDPNet, we compared it with several state-of-the-art methods on multi-center datasets. The results showed that, comparing against the suboptimal method, the DSC and HD95 of PDPNet were improved at least by 5.13% and 7.58% respectively on multi-center test sets. In addition, through ablations, we demonstrated that the proposed localization module can decrease the influence of normal tissues and therefore improve the generalization ability of the model. The weak semantic priors allow focusing on tumor regions to avoid missing small tumors and low-contrast tumors. The cross-scale correlation priors are beneficial for promoting the shape-aware ability for irregular tumors. Thus integrating them in a unified framework improved the multi-center breast tumor segmentation performance.

Three-dimensional (3D) radiative transfer (RT) models are frequently recognized as a prerequisite when using high spatial resolution remote sensing data in heterogeneous surfaces. However, most studies of 3D RT models have been restricted to limited applications due to the low computational efficiency. Therefore, this study proposed a graphic processing unit (GPU)-based solution for the ray tracing 3D RT model. A state-of-the-art graphics and compute application programming interface, Vulkan, was introduced to implement the RT process. A bounding box method was adopted for the computation acceleration. By comparison with a central processing unit (CPU)-based solution, the performance efficiency of the proposed solution is significantly better: the simulation time of a GPU model is significantly reduced by more than 99% when facing a large-scale simulation mission. The simulation accuracy of the two solutions is similar, with root mean squared errors (RMSEs) lower than 0.005, 0.032 and 0.31 K for the red, near-infrared (NIR) and brightness temperature images, respectively. An evaluation based on airborne multiangle measurements also indicated that the accuracy of the proposed solution was satisfactory for simulating the red and NIR bidirectional reflectance factor and brightness temperature directional anisotropies, with RMSEs lower than 0.003, 0.020 and 0.20 K, respectively, when treating the whole scene as a pixel. Considering the simulation accuracy and efficiency, a GPU-based model will be an important supplement to the CPU model.

Training convolutional neural networks (CNNs) for segmentation of pulmonary airway, artery, and vein is challenging due to sparse supervisory signals caused by the severe class imbalance between tubular targets and background. We present a CNNs-based method for accurate airway and artery-vein segmentation in non-contrast computed tomography. It enjoys superior sensitivity to tenuous peripheral bronchioles, arterioles, and venules. The method first uses a feature recalibration module to make the best use of features learned from the neural networks. Spatial information of features is properly integrated to retain relative priority of activated regions, which benefits the subsequent channel-wise recalibration. Then, attention distillation module is introduced to reinforce representation learning of tubular objects. Fine-grained details in high-resolution attention maps are passing down from one layer to its previous layer recursively to enrich context. Anatomy prior of lung context map and distance transform map is designed and incorporated for better artery-vein differentiation capacity. Extensive experiments demonstrated considerable performance gains brought by these components. Compared with state-of-the-art methods, our method extracted much more branches while maintaining competitive overall segmentation performance. Codes and models are available at http://www.pami.sjtu.edu.cn/News/56.

The use of diffusion tensor imaging (DTI) for studying the human heart in vivo is very challenging due to cardiac motion. This paper assesses the effects of cardiac motion on the human myocardial fiber architecture. To this end, a model for analyzing the effects of cardiac motion on signal intensity is presented. A Monte-Carlo simulation based on polarized light imaging data is then performed to calculate the diffusion signals obtained by the displacement of water molecules, which generate diffusion weighted (DW) images. Rician noise and in vivo motion data obtained from DENSE acquisition are added to the simulated cardiac DW images to produce motion-induced datasets. An algorithm based on principal components analysis filtering and temporal maximum intensity projection (PCATMIP) is used to compensate for motion-induced signal loss. Diffusion tensor parameters derived from motion-reduced DW images are compared to those derived from the original simulated DW images. Finally, to assess cardiac motion effects on in vivo fiber architecture, in vivo cardiac DTI data processed by PCATMIP are compared to those obtained from one trigger delay (TD) or one single phase acquisition. The results showed that cardiac motion produced overestimated fractional anisotropy and mean diffusivity as well as a narrower range of fiber angles. The combined use of shifted TD acquisitions and postprocessing based on image registration and PCATMIP effectively improved the quality of in vivo DW images and subsequently, the measurement accuracy of fiber architecture properties. This suggests new solutions to the problems associated with obtaining in vivo human myocardial fiber architecture properties in clinical conditions.