PAFE: A lightweight visualization-based fast malware classification method.

With the development of automated malware toolkits, cybersecurity faces evolving threats. Although visualization-based malware analysis has proven to be an effective method, existing approaches struggle with challenging malware samples due to alterations in the texture features of binary images during the visualization preprocessing stage, resulting in poor performance. Furthermore, to enhance classification accuracy, existing methods sacrifice prediction time by designing deeper neural network architectures. This paper proposes PAFE, a lightweight and visualization-based rapid malware classification method. It addresses the issue of texture feature variations in preprocessing through pixel-filling techniques and applies data augmentation to overcome the challenges of class imbalance in small sample datasets. PAFE combines multi-scale feature fusion and a channel attention mechanism, enhancing feature expression through modular design. Extensive experimental results demonstrate that PAFE outperforms the current state-of-the-art methods in both efficiency and effectiveness for malware variant classification, achieving an accuracy rate of 99.25 % with a prediction time of 10.04 ms.

Design of compensation algorithms for zero padding and its application to a patch based deep neural network.

In this article, compensation algorithms for zero padding are suggested to enhance the performance of deep convolutional neural networks. By considering the characteristics of convolving filters, the proposed methods efficiently compensate convolutional output errors due to zero padded inputs in a convolutional neural network. Primarily the algorithms are developed for patch based SRResNet for Single Image Super Resolution and the performance comparison is carried out using the SRResNet model but due to generalized nature of the padding algorithms its efficacy is tested in U-Net for Lung CT Image Segmentation. The proposed algorithms show better performance than the existing algorithm called partial convolution based padding (PCP), developed recently.

Split_ Composite: A Radar Target Recognition Method on FFT Convolution Acceleration.

Synthetic Aperture Radar (SAR) is renowned for its all-weather and all-time imaging capabilities, making it invaluable for ship target recognition. Despite the advancements in deep learning models, the efficiency of Convolutional Neural Networks (CNNs) in the frequency domain is often constrained by memory limitations and the stringent real-time requirements of embedded systems. To surmount these obstacles, we introduce the Split_ Composite method, an innovative convolution acceleration technique grounded in Fast Fourier Transform (FFT). This method employs input block decomposition and a composite zero-padding approach to streamline memory bandwidth and computational complexity via optimized frequency-domain convolution and image reconstruction. By capitalizing on FFT's inherent periodicity to augment frequency resolution, Split_ Composite facilitates weight sharing, curtailing both memory access and computational demands. Our experiments, conducted using the OpenSARShip-4 dataset, confirm that the Split_ Composite method upholds high recognition precision while markedly enhancing inference velocity, especially in the realm of large-scale data processing, thereby exhibiting exceptional scalability and efficiency. When juxtaposed with state-of-the-art convolution optimization technologies such as Winograd and TensorRT, Split_ Composite has demonstrated a significant lead in inference speed without compromising the precision of recognition.

Analysis of Tribological Properties of Hardfaced High-Chromium Layers Subjected to Wear in Abrasive Soil Mass.

This article presents the results of abrasion wear resistance tests of wear-resistant steel and surfacing under laboratory conditions and natural operation. Abrasion wear resistance determined on the basis of the study by determining geometrical characteristics of the alloying additives using computer image analysis methods, as well as examining the changes occurring on the surface of the workpieces and their wear intensity. Based on the results obtained from laboratory tests, it was noted that AR steel exhibited 14 times greater wear than the padding weld. This wear is affected by alloy additives, which, for the padding weld, are chromium additives. The microstructure image shows that soil mass had a destructive effect mainly on the matrix of the material, whereas in the areas with high concentrations of chromium precipitates, this effect was significantly weaker. The operational test results showed that within the area of the tine subjected to hardfacing, the material loss was lower than that for the same area of the tine in the as-delivered state. For the hardfaced tine, a 7% loss of volume was noted in relation to the operating part before testing and following the friction process. However, for the operating part in the as-delivered state, this difference amounted to 12%.

Use of Heat-Applied Coatings to Reduce Wear on Agricultural Machinery Components.

This article presents the effect of the conditions of abrasive compounds on the wear of samples made by different methods. The 28MnB5 steel was used, which is intended for agricultural components, to which two arc and laser coatings were applied. The study included the analysis of microstructure, microhardness, roughness, and tribological experiments on a dedicated stand. The arc coating was found to significantly improve the tribological properties compared to the samples without the coating. Varied wear results were obtained for the laser coating depending on the parameters of the abrasive compound. Studies of the surface roughness of the samples showed that the concentration and pH of the abrasives have a significant effect on the changes in the surface parameters after the tribological tests. The results of the tribological experiments indicated that wear resistance for some of the abrasive mass conditions was improved by the application of heat-applied coatings. In addition, it was found that the power consumption on the stand was the highest for abrasive mass conditions of a 10% moisture content and a pH of 10. For these test conditions, the mass loss was four times higher than for the parameter with W0% and pH7. The energy consumption of the stand was 60 kWh lower for this variant than for the parameter with W10% and pH10. The results of the study have important practical applications that can help in the selection of materials for agricultural machinery components, depending on the abrasive mass conditions.

The effects of postural support padding modifications to child restraints for children with disability on crash protection.

OBJECTIVE: Many children with physical disabilities need additional postural support when sitting and supplementary padding is used on standards approved child restraints to achieve this when traveling in a motor vehicle. However, the effect of this padding on crash protection for a child is unknown. This study aimed to investigate the effect of additional padding for postural support on crash protection for child occupants in forward facing child restraints. METHODS: Forty frontal sled tests at 49 km/h were conducted to compare Q1 anthropometric test device (ATD) responses in a forward-facing restraint, with and without additional padding in locations to increase recline of the restraint, and/or support the head, trunk and pelvis. Three padding materials were tested: cloth toweling, soft foam, and expanded polystyrene (EPS). The influence of padding on head excursion, peak 3 ms head acceleration, HIC15, peak 3 ms chest acceleration and chest deflection were analyzed. RESULTS: The influence of padding varied depending on the location of use. Padding used under the restraint to increase the recline angle increased head injury metrics. Toweling in multiple locations which included behind the head increased head excursion and chest injury metrics. There was minimal effect on injury risk measures with additional padding to support the sides of the head or the pelvis position. Rigid EPS foam, as recommended in Australian standards and guidelines, had minimal effect on injury metrics when used inside the restraint, as did tightly rolled or folded toweling secured to the restraint at single locations around the body of the child. CONCLUSIONS: This study does not support the use of postural support padding to increase recline of a forward-facing restraint or padding behind the head. Recommendations in published standards and guidelines to not use foam that is spongy, soft or easily compressed, with preference for secured firm foam or short-term use of tightly rolled or folded toweling under the child restraint cover is supported. This study also highlights the importance of considering the whole context of child occupant protection when using additional padding, particularly the change in the child's seated position when adding padding in relation to the standard safety features of the restraint.

Allergic Contact Dermatitis in the Right Forearm Following Splint Application for Distal Radius Fracture: A Rare Case of Plaster Cotton Allergy.

Allergic contact dermatitis (ACD) after splint or cast application (plaster of Paris) is infrequently encountered in orthopedic and traumatology clinical practice. This case study aims to elucidate the identification of ACD after splint application, highlight the conditions that warrant vigilance, and outline the precautions and optional treatment methods available in such instances. A 56-year-old right-hand dominant female presented to the emergency department after a fall on her right hand, manifesting pain, swelling, and tenderness without neurovascular injury. Radiographs revealed a distal radius fracture, leading to the application of a plaster of Paris splint. Within one day, she returned to the emergency department with severe itching and burning in the right arm. The splint was removed, and a dermatology consultation confirmed ACD due to undercast cotton padding. After splint removal, the patient's fracture treatment continued using a shoulder-arm sling until the lesion healed. Topical antihistamine ointment and oral corticosteroids were prescribed. Regular follow-up revealed the healing and union of the fracture by the fifth week, with minimal residual skin color changes. This case underscores the importance of prompt diagnosis and appropriate treatment in managing such occurrences. A key takeaway is the crucial need to schedule a follow-up appointment with the patient within one day of applying the cast or splint. Skin problems can emerge rather than neurovascular issues following casts or splints. Educating patients on warning signs, including skin irritation, neurovascular deficits, and symptoms of compartment syndrome, ensures the timely identification of significant issues. Healthcare practitioners should inquire about patients' histories of allergic skin reactions, taking a proactive approach to prevent ACD through early intervention and preventive measures.

Prevention of Seroma Formation and Its Sequelae After Axillary Lymph Node Dissection: An Up-to-Date Systematic Review and Guideline for Surgeons.

INTRODUCTION: Seroma formation after axillary lymph node dissection (ALND) remains a troublesome complication with significant morbidity. Numerous studies have tried to identify techniques to prevent seroma formation. The aim of this systematic review and network meta-analysis is to use available literature to identify the best intervention for prevention of seroma after standalone ALND. METHODS: A literature search was performed for all comparative articles regarding seroma formation in patients undergoing a standalone ALND or ALND with breast-conserving surgery in the last 25 years. Data regarding seroma formation, clinically significant seroma (CSS), surgical site infections (SSI), and hematomas were collected. The network meta-analysis was performed using a random effects model and the level of inconsistency was evaluated using the Bucher method. RESULTS: A total of 19 articles with 1962 patients were included. Ten different techniques to prevent seroma formation were described. When combining direct and indirect comparisons, axillary drainage until output is less than 50 ml per 24 h for two consecutive days results in significantly less CSS. The use of energy sealing devices, padding, tissue glue, or patches did not significantly reduce the incidence of CSS. When comparing the different techniques with regard to SSIs, no statistically significant differences were seen. CONCLUSIONS: To prevent CSS after ALND, axillary drainage is the most valuable and scientifically proven measure. On the basis of the results of this systematic review with network meta-analysis, removing the drain when output is < 50 ml per 24 h for two consecutive days irrespective of duration seems best. Since drainage policies vary widely, an evidence-based guideline is needed.

Sign Language Motion Generation from Sign Characteristics.

This paper proposes, analyzes, and evaluates a deep learning architecture based on transformers for generating sign language motion from sign phonemes (represented using HamNoSys: a notation system developed at the University of Hamburg). The sign phonemes provide information about sign characteristics like hand configuration, localization, or movements. The use of sign phonemes is crucial for generating sign motion with a high level of details (including finger extensions and flexions). The transformer-based approach also includes a stop detection module for predicting the end of the generation process. Both aspects, motion generation and stop detection, are evaluated in detail. For motion generation, the dynamic time warping distance is used to compute the similarity between two landmarks sequences (ground truth and generated). The stop detection module is evaluated considering detection accuracy and ROC (receiver operating characteristic) curves. The paper proposes and evaluates several strategies to obtain the system configuration with the best performance. These strategies include different padding strategies, interpolation approaches, and data augmentation techniques. The best configuration of a fully automatic system obtains an average DTW distance per frame of 0.1057 and an area under the ROC curve (AUC) higher than 0.94.

Dosimetric characterization of foam padding with posterior fields in palliative radiation therapy.

Patients undergoing external beam radiation therapy for the palliative treatment of painful bony metastases may have difficulty maintaining a still position on a rigid uncovered couch top, both during CT simulation as well as during patient setup, image guidance, and treatment on the linear accelerator. For these patients, a thin foam pad or mattress is sometimes used to mitigate patient discomfort. It was desired to quantify the effect of the padding in cases in which the patient is to be treated supine with posterior beams when the majority of the beam weighting traverses both the couch and the pad. Ion chamber measurements in-phantom were acquired with 6 MV, 10 MV, and 15 MV photon beams. At depths of maximum dose, the pad resulted in a difference of signal collected ≤1%. At the phantom surface, the pad resulted in an increase in signal ranging from 1% to 6.5% for the measured beams. CT data of the pad, both with and without applied pressure, indicated that the pad had average HU values close to air.

A bio-inspired positional embedding network for transformer-based models.

Owing to the progress of transformer-based networks, there have been significant improvements in the performance of vision models in recent years. However, there is further potential for improvement in positional embeddings that play a crucial role in distinguishing information across different positions. Based on the biological mechanisms of human visual pathways, we propose a positional embedding network that adaptively captures position information by modeling the dorsal pathway, which is responsible for spatial perception in human vision. Our proposed double-stream architecture leverages large zero-padding convolutions to learn local positional features and utilizes transformers to learn global features, effectively capturing the interaction between dorsal and ventral pathways. To evaluate the effectiveness of our method, we implemented experiments on various datasets, employing differentiated designs. Our statistical analysis demonstrates that the simple implementation significantly enhances image classification performance, and the observed trends demonstrate its biological plausibility.

B 1 + $$ {\mathrm{B}}_1

PURPOSE: Nuclear Overhauser effect magnetization transfer ratio (NOEMTR ) is a technique used to investigate brain lipids and macromolecules in greater detail than other techniques and benefits from increased contrast at 7 T. However, this contrast can become degraded because of B 1 + $$ {\mathrm{B}}_1^{+} $$ inhomogeneities present at ultra-high field strengths. High-permittivity dielectric pads (DP) have been used to correct for these inhomogeneities via displacement currents generating secondary magnetic fields. The purpose of this work is to demonstrate that dielectric pads can be used to mitigate B 1 + $$ {\mathrm{B}}_1^{+} $$ inhomogeneities and improve NOEMTR contrast in the temporal lobes at 7 T. METHODS: Partial 3D NOEMTR contrast images and whole brain B 1 + $$ {\mathrm{B}}_1^{+} $$ field maps were acquired on a 7 T MRI across six healthy subjects. Calcium titanate DP, having a relative permittivity of 110, was placed next to the subject's head near the temporal lobes. Pad corrected NOEMTR images had a separate postprocessing linear correction applied. RESULTS: DP provided supplemental B 1 + $$ {\mathrm{B}}_1^{+} $$ to the temporal lobes while also reducing the B 1 + $$ {\mathrm{B}}_1^{+} $$ magnitude across the posterior and superior regions of the brain. This resulted in a statistically significant increase in NOEMTR contrast in substructures of the temporal lobes both with and without linear correction. The padding also produced a convergence in NOEMTR contrast toward approximately equal mean values. CONCLUSION: NOEMTR images showed significant improvement in temporal lobe contrast when DP were used, which resulted from an increase in B 1 + $$ {\mathrm{B}}_1^{+} $$ homogeneity across the entire brain slab. DP-derived improvements in NOEMTR are expected to increase the robustness of the brain substructural measures both in healthy and pathological conditions.

Offloading effects of a removable cast walker with and without modification for diabetes-related foot ulceration: a plantar pressure study.

BACKGROUND: Removable cast walkers (RCWs), with or without modifications, are used to offload diabetes-related foot ulcers (DRFUs), however there is limited data relating to their offloading effects. This study aimed to quantify plantar pressure reductions with an RCW with and without modification for DRFUs. METHODS: This within-participant, repeated measures study included 16 participants with plantar neuropathic DRFUs. Walking peak plantar pressures at DRFU sites were measured for four conditions: post-operative boot (control condition), RCW alone, RCW with 20 mm of felt adhered to an orthosis, and RCW with 20 mm of felt adhered to the foot. RESULTS: Compared to the control condition, the greatest amount of peak plantar pressure reduction occurred with the RCW with felt adhered to the foot (83.1% reduction, p < .001). The RCW with felt adhered to the foot also offered greater peak plantar pressure reduction than the RCW alone (51.3%, p = .021) and the RCW with felt adhered to an orthosis (31.4%, p = .009). CONCLUSION: The largest offloading effect recorded was with the RCW with felt adhered to the foot. High-quality randomised trials are now needed to evaluate the effectiveness of this device for healing DRFUs.

CryoEM single particle reconstruction with a complex-valued particle stack.

Single particle reconstruction (SPR) in cryoEM is an image processing task with an elaborate hierarchy that starts with many very noisy multi-frame images. Efficient representation of the intermediary image structures is critical for keeping the calculations manageable. One such intermediary structure is called a particle stack and contains cut-out images of particles in square boxes of predefined size. The micrograph that is the source of the boxed images is usually corrected for motion between frames prior to particle stack creation. However, the contrast transfer function (CTF) or its Fourier Transform point spread function (PSF) are not considered at this step. Historically, the particle stack was intended for large particles and for a tighter PSF, which is characteristic of lower resolution data. The field now performs analyses of smaller particles and to higher resolution, and these conditions result in a broader PSF that requires larger padding and slower calculations to integrate information for each particle. Consequently, the approach to handling structures such as the particle stack should be reexamined to optimize data processing. Here we propose to use as a source image for the particle stack a complex-valued image, in which CTF correction is implicitly applied as a real component of the image. We can achieve it by applying an initial CTF correction to the entire micrograph first and perform box cutouts as a subsequent step. The final CTF correction that we refine and apply later has a very narrow PSF, and so cutting out particles from micrographs that were approximately corrected for CTF does not require extended buffering, i.e. the boxes during the analysis only have to be large enough to encompass the particle. The Fourier Transform of an exit-wave reconstruction creates an image that has complex values. This is a complex value image considered in real space, opposed to standard SPR data processing where complex numbers appear only in Fourier space. This extension of the micrograph concept provides multiple advantages because the particle box size can be small and calculations crucial for high resolution reconstruction such as Ewald sphere correction, aberration refinement, and particle-specific defocus refinement can be performed on the small box data.

Identification of phosphorylation site using S-padding strategy based convolutional neural network.

Purpose: Abnormal phosphorylation has been proved to associate with a variety of human diseases, and the identification of phosphorylation sites is one of the research hotspots in healthcare. The study of phosphorylation site prediction in deep learning models often introduces a variety of information, and the utilization of complex models limits the usage scenarios of the models. Methods: An enhanced deep learning method with S-padding strategy based on convolutional neural network is proposed in this paper. The S-padding strategy forms a three-dimensional matrix with extension information from original amino acid sequences, and a corresponding 2D-CNN model is designed to abstract the comprehensive features of phosphorylation site area in protein sequences. Results: The fivefold cross-validation experiments are conducted, and the results show the performance of the proposed method on human dataset can achieve an accuracy of 89.68 % on serine/threonine sites and 88.16 % on tyrosine sites, respectively. Furthermore, phosphorylation site prediction on different organisms obtains the accuracy, sensitivity, and specificity of over 0.85, indicating a potential capability on phosphorylation site prediction task. Comparison result with existing models shows that the proposed method obtains better performance on both accuracy and AUC value, and the proposed method can further improve performance with sufficient training data. Conclusion: This method enables proteome-wide predictions via models trained on a large amount of phosphorylation data, further exploiting the potential of protein phosphorylation site identification, and helping to provide insights into phosphorylation mechanisms.

Stroke technique in C1 canoe slalom: a simulation study.

Male C1 canoe slalom athletes traditionally used cross transitions to move their paddle to the other side of the boat and off-side strokes to paddle on their non-dominant side. Conversely, female athletes often use a switching transition and on-side strokes on their non-dominant side. The purpose of this study was to use a computer model to assess the relation between cross- or switching techniques, and the relative strength (symmetry) of non-dominant compared to dominant side strokes to race times in C1 canoe slalom. We created a forward dynamics model to predict race times using stroke forces (from an indoor ergometer), drag forces (measured on-water), and probability distributions for stroke and transition times (measured from international canoe slalom competitions). The main effects from an ANOVA (p<0.05) were (i) for a given transition number and strength symmetry the race times were faster when using cross-transitions than switch-transitions (ii) for a given strength symmetry the race times became slower as the number of switch transitions increased, but there was minimal effect of the number of cross-transitions, and (iii) the closer the strength of the strokes were between the dominant and non-dominant side (as symmetry factor approached 100 %), the faster the race times.

A Zero-Padding Frequency Domain Convolutional Neural Network for SSVEP Classification.

The brain-computer interface (BCI) of steady-state visual evoked potential (SSVEP) is one of the fundamental ways of human-computer communication. The main challenge is that there may be a nonlinear relationship between different SSVEP in other states. For improving the performance of SSVEP BCI, a novel CNN algorithm model is proposed in this study. Based on the discrete Fourier transform to calculate the signal's power spectral density (PSD), we perform zero-padding in the signal's time domain to improve its performance on the PSD and make it more refined. In this way, the frequency point interval in the PSD of the SSVEP is consistent with the minimum gap between the stimulation frequency. Combining the nonlinear transformation capabilities of CNN in deep learning, a zero-padding frequency domain convolutional neural network (ZPFDCNN) model is proposed. Extensive experiments based on the SSVEP dataset validate the effectiveness of our method. The study verifies that the proposed ZPFDCNN method can improve the effectiveness of the SSVEP-based high-speed BCI ITR. It has massive potential in the application of BCI.

Zero-Padding and Spatial Augmentation-Based Gas Sensor Node Optimization Approach in Resource-Constrained 6G-IoT Paradigm.

Ultra-low-power is a key performance indicator in 6G-IoT ecosystems. Sensor nodes in this eco-system are also capable of running light-weight artificial intelligence (AI) models. In this work, we have achieved high performance in a gas sensor system using Convolutional Neural Network (CNN) with a smaller number of gas sensor elements. We have identified redundant gas sensor elements in a gas sensor array and removed them to reduce the power consumption without significant deviation in the node's performance. The inevitable variation in the performance due to removing redundant sensor elements has been compensated using specialized data pre-processing (zero-padded virtual sensors and spatial augmentation) and CNN. The experiment is demonstrated to classify and quantify the four hazardous gases, viz., acetone, carbon tetrachloride, ethyl methyl ketone, and xylene. The performance of the unoptimized gas sensor array has been taken as a "baseline" to compare the performance of the optimized gas sensor array. Our proposed approach reduces the power consumption from 10 Watts to 5 Watts; classification performance sustained to 100 percent while quantification performance compensated up to a mean squared error (MSE) of 1.12 × 10-2. Thus, our power-efficient optimization paves the way to "computation on edge", even in the resource-constrained 6G-IoT paradigm.

Alternatives and preferences for materials in use for pressure ulcer prevention: An experiment-reinforced literature review.

Alleviation of localised, sustained tissue loads and microclimate management are the most critical performance criteria for materials in use for pressure ulcer prevention, such as in prophylactic dressings, padding or cushioning. These material performance criteria can be evaluated by calculating the extents of matching between the material stiffness (elastic modulus) and the thermal conductivity of the protective dressing, padding or cushioning with the corresponding properties of native skin, separately or in combination. Based on these bioengineering performance criteria, hydrocolloids, which are commonly used for prophylaxis of medical device-related pressure ulcers, exhibit poor stiffness matching with skin. In addition, there is remarkable variability in the modulus and thermal conductivity matching levels of different material types used for pressure ulcer prevention, however, it appears that among the materials tested, hydrogels provide the optimal matching with skin, followed by gels and silicone foams. The stiffness matching for hydrocolloids appears to be inferior even to that of gauze. This article provides quantitative performance criteria and metrics for these evaluations, and grades commonly used material types to biomechanically guide clinicians and industry with regards to the selection of dressings for pressure ulcer prevention, both due to bodyweight forces and as a result of applied medical devices.

TransDIR: Deformable imaging registration network based on transformer to improve the feature extraction ability.

PURPOSE: Imaging registration has a significant contribution to guide and support physicians in the process of decision-making for diagnosis, prognosis, and treatment. However, existing registration methods based on the convolutional neural network cannot extract global features effectively, which significantly influences registration performance. Moreover, the smoothness of the displacement vector field (DVF) fails to be ensured due to the miss folding penalty. METHODS: In order to capture abundant global information as well as local information, we have proposed a novel 3D deformable image registration network based on Transformer (TransDIR). In the encoding phase, the transformer with the atrous reduction attention block is designed to capture the long-distance dependencies that are crucial for extracting global information. A zero-padding position encoder is embedded into the transformer to capture the local information. In the decoding phase, an up-sampling module based on an attention mechanism is designed to increase the significance of ROIs. Because of adding folding penalty term into loss function, the smoothness of DVF is improved. RESULTS: Finally, we carried out experiments on OASIS, LPBA40, MGH10, and MM-WHS open datasets to validate the effectiveness of TransDIR. Compared with LapIRN, the DSC score is improved by 1.1% and 0.9% on OASIS and LPBA40, separately. In addition, compared with VoxelMorph, the DSC score is improved by 2.8% on the basis of the folding index decreased by hundreds of times on MM-WHS. CONCLUSIONS: The results show that the TransDIR achieves robust registration and promising generalizability compared with LapIRN and VoxelMorph.