Edge Loss Research Articles

Removing random noise and improving the resolution of seismic data using deep‐learning transformers

AbstractPost‐stack data are susceptible to noise interference and have low resolution, which impacts the accuracy and efficiency of subsequent seismic data interpretation. To address this issue, we propose a deep learning approach called Seis‐SUnet, which achieves simultaneous random noise suppression and super‐resolution reconstruction of seismic data. First, the Conv‐Swin‐Block is designed to utilize ordinary convolution and Swin transformer to capture the long‐distance dependencies in the spatial location of seismic data, enabling the network to comprehensively comprehend the overall structure of seismic data. Second, to address the problem of weakening the effective signal during network mapping, we use a hybrid training strategy of L1 loss, edge loss and multi‐scale structural similarity loss. The edge loss function directs the network training to focus more on the high‐frequency information at the edges of seismic data by amplifying the weight. Additionally, the verification of synthetic and field seismic datasets confirms that Seis‐SUnet can effectively improve the signal‐to‐noise ratio and resolution of seismic data. By comparing it with traditional methods and two deep learning reconstruction methods, experimental results demonstrate that Seis‐SUnet excels in removing random noise, preserving the continuity of rock layers and maintaining faults as well as being strong robustness.

SDAM: A dual attention mechanism for high-quality fusion of infrared and visible images.

Image fusion of infrared and visible images to obtain high-quality fusion images with prominent infrared targets has important applications in various engineering fields. However, current fusion processes encounter problems such as unclear texture details and imbalanced infrared targets and texture detailed information, which lead to information loss. To address these issues, this paper proposes a method for infrared and visible image fusion based on a specific dual-attention mechanism (SDAM). This method employs an end-to-end network structure, which includes the design of channel attention and spatial attention mechanisms. Through these mechanisms, the method can fully exploit the texture details in the visible images while preserving the salient information in the infrared images. Additionally, an optimized loss function is designed to combine content loss, edge loss, and structure loss to achieve better fusion effects. This approach can fully utilize the texture detailed information of visible images and prominent information in infrared images, while maintaining better brightness and contrast, which improves the visual effect of fusion images. Through conducted ablation experiments and comparative evaluations on public datasets, our research findings demonstrate that the SDAM method exhibits superior performance in both subjective and objective assessments compared to the current state-of-the-art fusion methods.

An Adaptive Vehicle Detection Model for Traffic Surveillance of Highway Tunnels Considering Luminance Intensity.

Vehicle detection is essential for road traffic surveillance and active safety management. Deep learning methods have recently shown robust feature extraction capabilities and achieved improved detection results. However, vehicle detection models often perform poorly under abnormal lighting conditions, especially in highway tunnels. We proposed an adaptive vehicle detection model that accounts for varying luminance intensities to address this issue. The model categorizes the image data into abnormal and normal luminance scenarios. We employ an improved CycleGAN with edge loss as the adaptive luminance adjustment module for abnormal luminance scenarios. This module adjusts the brightness of the images to a normal level through a generative network. Finally, YOLOv7 is utilized for vehicle detection. The experimental results demonstrate that our adaptive vehicle detection model effectively detects vehicles under abnormal luminance scenarios in highway tunnels. The improved CycleGAN can effectively mitigate edge generation distortion. Under abnormal luminance scenarios, our model achieved a 16.3% improvement in precision, a 1.7% improvement in recall, and a 9.8% improvement in mAP_0.5 compared to the original YOLOv7. Additionally, our adaptive luminance adjustment module is transferable and can enhance the detection accuracy of other vehicle detection models.

Performance study of CPV/T systems with different tracking methods

Abstract This article built the model of concentrated photovoltaic and thermal (CPV/T) systems under the double-axis tracking mode in the east, west, south, and north directions, as well as the single-axis tracking mode in the north and south directions. According to this model, the function of the CPV/T system could be researched. The article established 3D models of large-scale CPV/T systems in different tracking modes and used this model to study the occlusion problem of CPV/T systems. Taking into account cosine loss, edge loss, and occlusion, this paper established energy reception models for different trace modes of the CPV/T system. The energy reception model calculated that the annual energy reception rate of the CPV/T system in the single-axis tracking mode was not a patch on the double-axis tracking CPV/T system. Subsequently, a scaled CPV/T total energy mathematical model was established under these two tracking mods in the article. Combined with the non-steady energy model, the annual total photoelectricity and photothermal output of the CPV/T systems in the single-axis tracking mode and the double-axis tracking mode with an area of 1600 m2 in Xi’an were calculated. The photoelectricity and photothermal outputs of the double-axis tracking mode were 169216 kW·h and 856183 kW·h, respectively. The single-axis tracking node had an annual output of 126898 kW·h for photoelectricity and 637490 kW·h for photothermal.

Radiological segmentation of knee meniscus ultrasound images based on boundary constraints and multi-scale fusion network

Knee osteoarthritis (KOA) is one of the most popular joint diseases endangering human health because of its high incidence, disability, and younger onset. However, there is no suitable method for early diagnosis, evaluation, and treatment of KOA. In recent years, some clinical studies have found that ultrasound can detect early changes in KOA in advance, and the automatic segmentation of ultrasound images can achieve rapid and effective quantitative research on KOA. However, in ultrasound images, the soft tissue boundaries of the lesion are blurred, making lesion segmentation difficult. Although the U-Net family is one of the best networks for image segmentation, they still have defects such as blurred segmentation boundaries, distorted morphology, and insufficient accuracy when segmenting ultrasound images. To address this issue, we added attention, atrous spatial pyramid pooling (ASSP), and edge loss function terms into the Unet3+ network, which improved the contour clarity and accuracy of output images (the improved Unet3+ Dice acc = 78.74%. Then, we extract the key features of the improved Unet3+ for outputting meniscus images: calculating meniscus area and distance, where the average accuracy of the area is: area_ avg_ acc = 91.12%, with an average distance accuracy of distance_ avg_ acc = 91.14%. This thesis creates a new dataset collection from West China Hospital, Sichuan University, and automated measurement of knee meniscus protrusion area has been achieved for the first time. This article is the first to apply deep learning to ultrasound image segmentation of the knee meniscus, helping doctors conduct qualitative and quantitative analysis of the diagnosis and treatment of early KOA. The results indicate that the improved Unet3+ can assist doctors in automatically diagnosing and evaluating KOA based on meniscus ultrasound images, which is beneficial for guiding early clinical intervention.

Prior image-based generative adversarial learning for multi-material decomposition in photon counting computed tomography

BackgroundPhoton counting detector computed tomography (PCD-CT) is a novel promising technique providing higher spatial resolution, lower radiation dose and greater energy spectrum differentiation, which create more possibilities to improve image quality. Multi-material decomposition is an attractive application for PCD-CT to identify complicated materials and provide accurate quantitative analysis. However, limited by the finite photon counting rate in each energy window of photon counting detector, the noise problem hinders the decomposition of high-quality basis material images. MethodsTo address this issue, an end-to-end multi-material decomposition network based on prior images is proposed in this paper. First, the reconstructed images corresponding to the full spectrum with less noise are introduced as prior information to improve the overall signal-to-noise ratio of the data. Then, a generative adversarial network is designed to mine the relationship between reconstructed images and basis material images based on the information interaction of material decomposition. Furthermore, a weighted edge loss is introduced to adapt to the structural differences of different basis material images. ResultsTo verify the performance of the proposed method, simulation and real studies are carried out. In simulation study of structured fibro-glandular tissue model, the results show that the proposed method decreased the root mean square error by 67 % and 26 % on adipose, 66 % and 28 % on fibroglandular, 52 % and 8 % on calcification, compared to butterfly network and dual interactive Wasserstein generative adversarial network. ConclusionExperimentally, the proposed method shows certain advantages over other methods on noise suppression effect, detail retention ability and decomposition accuracy.

Exploiting Optical Flow Guidance for Transformer-Based Video Inpainting.

Transformers have been widely used for video processing owing to the multi-head self attention (MHSA) mechanism. However, the MHSA mechanism encounters an intrinsic difficulty for video inpainting, since the features associated with the corrupted regions are degraded and incur inaccurate self attention. This problem, termed query degradation, may be mitigated by first completing optical flows and then using the flows to guide the self attention, which was verified in our previous work - flow-guided transformer (FGT). We further exploit the flow guidance and propose FGT++ to pursue more effective and efficient video inpainting. First, we design a lightweight flow completion network by using local aggregation and edge loss. Second, to address the query degradation, we propose a flow guidance feature integration module, which uses the motion discrepancy to enhance the features, together with a flow-guided feature propagation module that warps the features according to the flows. Third, we decouple the transformer along the temporal and spatial dimensions, where flows are used to select the tokens through a temporally deformable MHSA mechanism, and global tokens are combined with the inner-window local tokens through a dual-perspective MHSA mechanism. FGT++ is experimentally evaluated to be outperforming the existing video inpainting networks qualitatively and quantitatively.

HD-Former: A hierarchical dependency Transformer for medical image segmentation

Medical image segmentation is a compelling fundamental problem and an important auxiliary tool for clinical applications. Recently, the Transformer model has emerged as a valuable tool for addressing the limitations of convolutional neural networks by effectively capturing global relationships and numerous hybrid architectures combining convolutional neural networks (CNNs) and Transformer have been devised to enhance segmentation performance. However, they suffer from multilevel semantic feature gaps and fail to account for multilevel dependencies between space and channel. In this paper, we propose a hierarchical dependency Transformer for medical image segmentation, named HD-Former. First, we utilize a Compressed Bottleneck (CB) module to enrich shallow features and localize the target region. We then introduce the Dual Cross Attention Transformer (DCAT) module to fuse multilevel features and bridge the feature gap. In addition, we design the broad exploration network (BEN) that cascades convolution and self-attention from different percepts to capture hierarchical dense contextual semantic features locally and globally. Finally, we exploit uncertain multitask edge loss to adaptively map predictions to a consistent feature space, which can optimize segmentation edges. The extensive experiments conducted on medical image segmentation from ISIC, LiTS, Kvasir-SEG, and CVC-ClinicDB datasets demonstrate that our HD-Former surpasses the state-of-the-art methods in terms of both subjective visual performance and objective evaluation. Code: https://github.com/barcelonacontrol/HD-Former.

An improvement method for pancreas CT segmentation using superpixel-based active contour

Objective. Pancreas is one of the most challenging organs for Computed Tomograph (CT) image automatic segmentation due to its complex shapes and fuzzy edges. It is simple and universal to use the traditional segmentation method as a post-processor of deep learning method for segmentation accuracy improvement. As the most suitable traditional segmentation method for pancreatic segmentation, the active contour model (ACM), still suffers from the problems of weak boundary leakage and slow contour evolution speed. Therefore, a convenient post-processor for any deep learning methods using superpixel-based active contour model (SbACM) is proposed to improve the segmentation accuracy. Approach. Firstly, the superpixels with strong adhesion to edges are used to guide the design of narrowband and energy function. A multi-scale evolution strategy is also proposed to reduce the weak boundary leakage and comprehensively improve the evolution speed. Secondly, using the original image and the coarse segmentation results obtained from deep learning methods as inputs, the proposed SbACM method is used as a post-processor for fine segmentation. Finally, the pancreatic segmentation public dataset TCIA from the National Institutes of Health(NIH, USA) is used for evaluation, and the Wilcoxon Test confirmed that the improvement of proposed method is statistically significant. Main results. (1) the superpixel-based narrowband shape and dynamic edge energy of the proposed SbACM work for boundary leakage reduction, as well as the multi-scale evolution strategy and dynamic narrowband width for the evolution speed improvement; (2) as a post-processor, SbACM can increase the Dice similarity coefficients (DSC) of five typical UNet-based models, including UNet, SS-UNet, PBR UNet, ResDSN, and nnUNet, 2.35% in average and 9.04% in maximum. (3) Based on the best backbone nnUNet, the proposed post-processor performs better than either adding edge awareness or adding edge loss in segmentation enhancement without increasing the complexity and training time of deep learning models. Significance. The proposed SbACM can improve segmentation accuracy with the lowest cost, especially in cases of squeezed fuzzy edges with similar neighborhood , and complex edges.

Enhancing scanning electron microscopy imaging quality of weakly conductive samples through unsupervised learning

Scanning electron microscopy (SEM) is a crucial tool for analyzing submicron-scale structures. However, the attainment of high-quality SEM images is contingent upon the high conductivity of the material due to constraints imposed by its imaging principles. For weakly conductive materials or structures induced by intrinsic properties or organic doping, the SEM imaging quality is significantly compromised, thereby impeding the accuracy of subsequent structure-related analyses. Moreover, the unavailability of paired high–low quality images in this context renders the supervised-based image processing methods ineffective in addressing this challenge. Here, an unsupervised method based on Cycle-consistent Generative Adversarial Network (CycleGAN) was proposed to enhance the quality of SEM images for weakly conductive samples. The unsupervised model can perform end-to-end learning using unpaired blurred and clear SEM images from weakly and well-conductive samples, respectively. To address the requirements of material structure analysis, an edge loss function was further introduced to recover finer details in the network-generated images. Various quantitative evaluations substantiate the efficacy of the proposed method in SEM image quality improvement with better performance than the traditional methods. Our framework broadens the application of artificial intelligence in materials analysis, holding significant implications in fields such as materials science and image restoration.

Emb-trattunet: a novel edge loss function and transformer-CNN architecture for multi-classes pneumonia infection segmentation in low annotation regimes

One of the primary challenges in applying deep learning approaches to medical imaging is the limited availability of data due to various factors. These factors include concerns about data privacy and the requirement for expert radiologists to perform the time-consuming and labor-intensive task of labeling data, particularly for tasks such as segmentation. Consequently, there is a critical need to develop novel approaches for few-shot learning tasks in this domain. In this work, we propose a Novel CNN-Transformer Fusion scheme to segment Multi-classes pneumonia infection from limited CT-scans data. In total, there are three main contributions: (i) CNN-Transformer encoders fusion, which allows to extract and fuse richer features in the encoding phase, which contains: local, global and long-range dependencies features, (ii) Multi-Branches Skip Connection (MBSC) is proposed to extract and fuse richer features from the encoder features then integrate them into the decoder layers, where MBSC blocks extract higher-level features related to the finer details of different infection types, and (iii) a Multi-classes Boundary Aware Cross-Entropy (MBA-CE) Loss function is proposed to deal with fuzzy boundaries, enhance the separability between classes and give more attention to the minority classes. The performance of the proposed approach is evaluated using two evaluation scenarios and compared with different baseline and state-of-the-art segmentation architectures for Multi-classes Covid-19 segmentation. The obtained results show that our approach outperforms the comparison methods in both Ground-Glass Opacity (GGO) and Consolidation segmentation. On the other hand, our approach shows consistent performance when the training data is reduced to half, which proves the efficiency of our approach in few-shot learning. In contrast, the performance of the comparison methods drops in this scenario. Moreover, our approach is able to deal with imbalanced data classes. These advantages prove the effectiveness and efficiency of the proposed EMB-TrAttUnet approach in a pandemic scenario where time is critical to save patient lives.

Internal and edge-effect environmental monitoring to assess the threats facing intact tropical peat swamp forest in Central Kalimantan

Peat swamp forest (PSF) provides vital environmental benefits, including carbon storage. However, almost half of the Indonesia’s PSF areas are extensively degraded. Internal threats such as drainage contribute to peat drying, vegetation damage and heightened fire risk. Equally, forest edges are threatened from human activity and fire spread, leading to forest edge regression. This study assesses biophysical changes within an area of intact tropical PSF and along their edges in Central Kalimantan. To assess these threats, we established seven transects perpendicular to the forest edge, transitioning from degraded to intact forest. Forty-nine dipwells, and three loggers were installed to assess drainage and subsidence. We established seed traps and seedling plots along three transects at three distances from the forest edge (three replicates per distance) to assess forest edge regeneration potential. Furthermore, we are analysing historic remote sensing data from the last two decades to study land cover change and forest edge loss related to fire. Equipment was established February 2022, with data collection underway for 12 months. Here, we present our initial findings. The results will provide quantitative and qualitative data analysis to describe potential threat levels facing this forest and give recommendations for developing a targeted conservation action plan.

Uncertainty quantification of geometric variations in transonic high-pressure turbine with different trailing edge and throat area distribution

The transonic high-pressure turbine is a critical component in modern aviation engines which is susceptible to geometric variations throughout its entire lifecycle. However, the specific impact and underlying mechanism of geometric deviations on the performance are not yet fully understood. This article presents a framework for modeling, quantifying, and analyzing geometric variations in typical transonic high-pressure turbines. It aims to investigate the mechanism by which geometric variations impact aerodynamic performance across the entire operating range. Additionally, the study explores potential improvement methods for controlling the adverse effects caused by geometric deviations. The results indicate that within the entire operating range, the trailing edge and throat have the greatest impact on the fluctuation of total pressure loss. Deviation in the trailing edge and throat directly impacts the range of the base region and base pressure, leading to changes of shock waves and wake flow. This causes fluctuations in shock wave loss and trailing edge loss, ultimately resulting in the fluctuation of total pressure loss. Subsequently, two methods were proposed to suppress the influence of geometric deviations: the utilization of an elliptical trailing edge and a specially designed throat profile. These approaches effectively reduce the loss fluctuations caused by geometric deviations, particularly under typical transonic conditions.

Paint-CUT: A Generative Model for Chinese Landscape Painting Based on Shuffle Attentional Residual Block and Edge Enhancement

As one of the precious cultural heritages, Chinese landscape painting has developed unique styles and techniques. Researching the intelligent generation of Chinese landscape paintings from photos can benefit the inheritance of traditional Chinese culture. To address detail loss, blurred outlines, and poor style transfer in present generated results, a model for generating Chinese landscape paintings from photos named Paint-CUT is proposed. In order to solve the problem of detail loss, the SA-ResBlock module is proposed by combining shuffle attention with the resblocks in the generator, which is used to enhance the generator’s ability to extract the main scene information and texture features. In order to solve the problem of poor style transfer, perceptual loss is introduced to constrain the model in terms of content and style. The pre-trained VGG is used to extract the content and style features to calculate the perceptual loss and, then, the loss can guide the model to generate landscape paintings with similar content to landscape photos and a similar style to target landscape paintings. In order to solve the problem of blurred outlines in generated landscape paintings, edge loss is proposed to the model. The Canny edge detection is used to generate edge maps and, then, the edge loss between edge maps of landscape photos and generated landscape paintings is calculated. The generated landscape paintings have clear outlines and details by adding edge loss. Comparison experiments and ablation experiments are performed on the proposed model. Experiments show that the proposed model can generate Chinese landscape paintings with clear outlines, rich details, and realistic style. Generated paintings not only retain the details of landscape photos, such as texture and outlines of mountains, but also have similar styles to the target paintings, such as colors and brush strokes. So, the generation quality of Chinese landscape paintings has improved.

Cross2SynNet: cross-device-cross-modal synthesis of routine brain MRI sequences from CT with brain lesion.

CT and MR are often needed to determine the location and extent of brain lesions collectively to improve diagnosis. However, patients with acute brain diseases cannot complete the MRI examination within a short time. The aim of the study is to devise a cross-device and cross-modal medical image synthesis (MIS) method Cross2SynNet for synthesizing routine brain MRI sequences of T1WI, T2WI, FLAIR, and DWI from CT with stroke and brain tumors. For the retrospective study, the participants covered four different diseases of cerebral ischemic stroke (CIS-cohort), cerebral hemorrhage (CH-cohort), meningioma (M-cohort), glioma (G-cohort). The MIS model Cross2SynNet was established on the basic architecture of conditional generative adversarial network (CGAN), of which, the fully convolutional Transformer (FCT) module was adopted into generator to capture the short- and long-range dependencies between healthy and pathological tissues, and the edge loss function was to minimize the difference in gradient magnitude between synthetic image and ground truth. Three metrics of mean square error (MSE), peak signal-to-noise ratio (PSNR), and structure similarity index measure (SSIM) were used for evaluation. A total of 230 participants (mean patient age, 59.77years ± 13.63 [standard deviation]; 163 men [71%] and 67 women [29%]) were included, including CIS-cohort (95 participants between Dec 2019 and Feb 2022), CH-cohort (69 participants between Jan 2020 and Dec 2021), M-cohort (40 participants between Sep 2018 and Dec 2021), and G-cohort (26 participants between Sep 2019 and Dec 2021). The Cross2SynNet achieved averaged values of MSE = 0.008, PSNR = 21.728, and SSIM = 0.758 when synthesizing MRIs from CT, outperforming the CycleGAN, pix2pix, RegGAN, Pix2PixHD, and ResViT. The Cross2SynNet could synthesize the brain lesion on pseudo DWI even if the CT image did not exhibit clear signal in the acute ischemic stroke patients. Cross2SynNet could achieve routine brain MRI synthesis of T1WI, T2WI, FLAIR, and DWI from CT with promising performance given the brain lesion of stroke and brain tumor.

Improving polygon image segmentation by enhancing U-Net architecture

The crucial task of polyp recognition in medical imaging plays a pivotal role in the early detection and prevention of colorectal cancer. Semantic segmentation, particularly utilizing sophisticated deep learning models such as U-Net, has demonstrated promising results in the realm of polyp segmentation. However, the traditional U-Net structure sometimes grapples with accurately delineating the edges of polyps, which subsequently impacts the overall performance of segmentation. To address this issue, the current study introduces a novel approach by proposing a modified framework of U-Net, equipped with an enhanced edge loss function. This function is designed to ameliorate the accuracy of segmentation within polyp images. The aim is to elevate the model’s capacity to capture intricate details, specifically the edges, which is an area where standard U-Net structures often falter. Experimental outcomes of this study serve to underscore the effectiveness of the proposed approach in accomplishing superior segmentation of edges and improved overall performance in polyp recognition. By successfully tackling the challenges inherent to polyp edge segmentation, the modified U-Net model contributes significantly towards more precise diagnostic systems in the field of medical imaging. Consequently, this research is poised to make a valuable contribution to advancements in the prevention and early detection of colorectal cancer.

Application value of bone cement containing rhbFGF and RHBMP-2 in PKP treatment of osteoporotic lumbar compression fracture

To investigate the effect of bone cement containing recombinant human basic fibroblast growth factor (rhbFGF) and recombinant human bone morphogenetic protein-2 (rhBMP-2) in percutaneous kyphoplasty(PKP)treatment of osteoporotic vertebral compression fracture(OVCF). A total of 103 OVCF patients who underwent PKP from January 2018 to January 2021 were retrospectively analyzed, including 40 males and 63 females, aged from 61 to 78 years old with an average of (65.72±3.29) years old. The injury mechanism included slipping 33 patients, falling 42 patients, and lifting injury 28 patients. The patients were divided into three groups according to the filling of bone cement. Calcium phosphate consisted of 34 patients, aged(65.1±3.3) years old, 14 males and 20 females, who were filled with calcium phosphate bone cement. rhBMP-2 consisted of 34 patients, aged (64.8±3.2) years old, 12 males and 22 females, who were filled with bone cement containing rhBMP-2. And rhbFGF+rhBMP-2 consisted of 35 patients, aged (65.1±3.6) years old, 14 males and 21 females, who were filled with bone cement containing rhbFGF and rhBMP-2. Oswestry disability index (ODI), bone mineral density, anterior edge loss height, anterior edge compression rate of injured vertebra, visual analog scale (VAS) of pain, and the incidence of refracture were compared between groups. All patients were followed for 12 months. Postoperative ODI and VAS score of the three groups decreased (P<0.001), while bone mineral density increased (P<0.001), anterior edge loss height, anterior edge compression rate of injured vertebra decreased first and then slowly increased (P<0.001). ODI and VAS of group calcium phosphate after 1 months, 6 months, 12 months were lower than that of rhBMP-2 and group rhbFGF+rhBMP-2(P<0.05), bone mineral density after 6 months, 12 months was higher than that of rhBMP-2 and group calcium phosphate(P<0.05), and anterior edge loss height, anterior edge compression rate of injured vertebra of group rhbFGF+rhBMP-2 after 6 months and 12 months were lower than that of group rhBMP-2 and group calcium phosphate(P<0.05). There was no statistical difference in the incidence of re-fracture among the three groups (P>0.05). Bone cement containing rhbFGF and rhBMP-2 could more effectively increase bone mineral density in patients with OVCF, obtain satisfactory clinical and radiological effects after operation, and significantly improve clinical symptoms.

Combining residual structure and edge loss for face image restoration with generative adversarial networks

Combining residual structure and edge loss for face image restoration with generative adversarial networks

Combined Edge Loss UNet for Optimized Segmentation in Total Knee Arthroplasty Preoperative Planning.

Bone segmentation and 3D reconstruction are crucial for total knee arthroplasty (TKA) surgical planning with Personalized Surgical Instruments (PSIs). Traditional semi-automatic approaches are time-consuming and operator-dependent, although they provide reliable outcomes. Moreover, the recent expansion of artificial intelligence (AI) tools towards various medical domains is transforming modern healthcare. Accordingly, this study introduces an automated AI-based pipeline to replace the current operator-based tibia and femur 3D reconstruction procedure enhancing TKA preoperative planning. Leveraging an 822 CT image dataset, a novel patch-based method and an improved segmentation label generation algorithm were coupled to a Combined Edge Loss UNet (CEL-UNet), a novel CNN architecture featuring an additional decoding branch to boost the bone boundary segmentation. Root Mean Squared Errors and Hausdorff distances compared the predicted surfaces to the reference bones showing median and interquartile values of 0.26 (0.19-0.36) mm and 0.24 (0.18-0.32) mm, and of 1.06 (0.73-2.15) mm and 1.43 (0.82-2.86) mm for the tibia and femur, respectively, outperforming previous results of our group, state-of-the-art, and UNet models. A feasibility analysis for a PSI-based surgical plan revealed sub-millimetric distance errors and sub-angular alignment uncertainties in the PSI contact areas and the two cutting planes. Finally, operational environment testing underscored the pipeline's efficiency. More than half of the processed cases complied with the PSI prototyping requirements, reducing the overall time from 35 min to 13.1 s, while the remaining ones underwent a manual refinement step to achieve such PSI requirements, performing the procedure four to eleven times faster than the manufacturer standards. To conclude, this research advocates the need for real-world applicability and optimization of AI solutions in orthopedic surgical practice.

A full-scale hierarchical encoder-decoder network with cascading edge-prior for infrared and visible image fusion

Most existing deep learning-based infrared and visible image fusion methods always fail to consider the full-scale long-range correlation and the prior knowledge, resulting in the fused images with low-contrast salient objects and blurred edge details. To overcome these drawbacks, a full-scale hierarchical encoder-decoder network with cascading edge-prior for infrared and visible image fusion is proposed. First, a top-down encoder extracts the hierarchical representations from source image. Then, to inject edge priors into the network and capture the progressive semantic correlations, a triple fusion mechanism is proposed including edge image fusion based on maximum fusion strategy, single-scale shallow layer fusion and full-scale semantic layer fusion based on dual-attention fusion (DAF) strategy. The fused full-scale semantic features (F2SF) are obtained by capturing the long-range affinities of the full-scale. At the same time, a cascading edge-prior branch (CEPB) is designed to embed the fused edge knowledge into fused single-scale shallow features, jointly guiding the decoder to focus on abundant details layer-by-layer on the basis of F2SF, thus recovering the edge and texture details of the fused image well. Finally, a novel loss function consisting of SSIM, intensity and edge loss is constructed to further maintain the network with better edge representation and reconstruction capability. Compared with existing state-of-the-art fusion methods, the proposed method has better performance in terms of both visual evaluation and objective evaluation on public datasets. The source code is available at https://github.com/lxq-jnu/FSFusion.