Application of one-dimensional hierarchical network assisted screening for cervical cancer based on Raman spectroscopy combined with attention mechanism

Cervical cancer is one of the most common malignant tumors among women, and its pathological change is a relatively slow process. If it can be detected in time and treated properly, it can effectively reduce the incidence rate and mortality rate of cervical cancer, so the early screening of cervical cancer is particularly critical and significant. In this paper, we used Raman spectroscopy technology to collect the tissue sample data of patients with cervicitis, Low-grade Squamous Intraepithelial Lesion, High-grade Squamous Intraepithelial Lesion, Well differentiated squamous cell carcinoma, Moderately differentiated squamous cell carcinoma, Poorly differentiated squamous cell carcinoma and cervical adenocarcinoma. A one-dimensional hierarchical convolutional neural network based on attention mechanism was constructed to classify and identify seven types of tissue samples. The attention mechanism Efficient Channel Attention Networks module and Squeeze-and-Excitation Networks module were combined with the established one-dimensional convolutional hierarchical network model, and the results showed that the combined model had better diagnostic performance. The average accuracy, F1, and AUC of the Principal Component Analysis-Squeeze and Excitation-hierarchical network model after 5-fold cross validations could reach 96.49%±2.12%, 0.9663±0.0253, and 0.9815±0.0224, respectively, which were 1.58%, 0.0140, and 0.008 higher than those of hierarchical network. The recall rate of the Principal Component Analysis-Efficient Channel Attention-hierarchical network model was as high as 96.78%±2.85%, which is 1.47% higher than hierarchical network. Compared with the classification results of traditional CNN and ResNet for seven types of cervical cancer staging, the accuracy of the Principal Component Analysis-Squeeze and Excitation-hierarchical network model is 3.33% and 11.05% higher, respectively. The experimental results indicate that the model established in this study is easy to operate and has high accuracy. It has good reference value for rapid screening of cervical cancer, laying a foundation for further research on Raman spectroscopy as a clinical diagnostic method for cervical cancer.

Multi-branch myocardial infarction detection and localization framework based on multi-instance learning and domain knowledge

OBJECTIVE

Myocardial infarction (MI) is a serious cardiovascular disease that can cause irreversible damage to the heart, making early identification and treatment crucial. However, automatic MI detection and localization from an electrocardiogram (ECG) remain challenging. In this study, we propose two models, MFB-SENET and MFB-DMIL, for MI detection and localization, respectively.

APPROACH

The MFB-SENET model is designed to detect MI, while the MFB-DMIL model is designed to localize MI. The MI localization model employs a specialized attention mechanism to integrate multi-instance learning with domain knowledge. This approach incorporates handcrafted features and introduces a new loss function called lead-loss, to improve MI localization. Grad-CAM is employed to visualize the decision-making process. Main Results: The proposed method was evaluated on the PTB and PTB-XL databases. Under the inter-patient scheme, the accuracy of MI detection and localization on the PTB database reached 93.88% and 67.17%, respectively. The accuracy of MI detection and localization on the PTB-XL database were 94.89% and 85.83%, respectively.

SIGNIFICANCE

Our method achieved comparable or better performance than other state-of-the-art algorithms. The proposed method combined deep learning and medical domain knowledge, demonstrates effectiveness and reliability, holding promise as an efficient MI diagnostic tool to assist physicians in formulating accurate diagnoses. .

Tomato leaf disease detection based on attention mechanism and multi-scale feature fusion

When detecting tomato leaf diseases in natural environments, factors such as changes in lighting, occlusion, and the small size of leaf lesions pose challenges to detection accuracy. Therefore, this study proposes a tomato leaf disease detection method based on attention mechanisms and multi-scale feature fusion. Firstly, the Convolutional Block Attention Module (CBAM) is introduced into the backbone feature extraction network to enhance the ability to extract lesion features and suppress the effects of environmental interference. Secondly, shallow feature maps are introduced into the re-parameterized generalized feature pyramid network (RepGFPN), constructing a new multi-scale re-parameterized generalized feature fusion module (BiRepGFPN) to enhance feature fusion expression and improve the localization ability for small lesion features. Finally, the BiRepGFPN replaces the Path Aggregation Feature Pyramid Network (PAFPN) in the YOLOv6 model to achieve effective fusion of deep semantic and shallow spatial information. Experimental results indicate that, when evaluated on the publicly available PlantDoc dataset, the model's mean average precision (mAP) showed improvements of 7.7%, 11.8%, 3.4%, 5.7%, 4.3%, and 2.6% compared to YOLOX, YOLOv5, YOLOv6, YOLOv6-s, YOLOv7, and YOLOv8, respectively. When evaluated on the tomato leaf disease dataset, the model demonstrated a precision of 92.9%, a recall rate of 95.2%, an F1 score of 94.0%, and a mean average precision (mAP) of 93.8%, showing improvements of 2.3%, 4.0%, 3.1%, and 2.7% respectively compared to the baseline model. These results indicate that the proposed detection method possesses significant detection performance and generalization capabilities.

ESF-YOLO: an accurate and universal object detector based on neural networks

As an excellent single-stage object detector based on neural networks, YOLOv5 has found extensive applications in the industrial domain; however, it still exhibits certain design limitations. To address these issues, this paper proposes Efficient Scale Fusion YOLO (ESF-YOLO). Firstly, the Multi-Sampling Conv Module (MSCM) is designed, which enhances the backbone network's learning capability for low-level features through multi-scale receptive fields and cross-scale feature fusion. Secondly, to tackle occlusion issues, a new Block-wise Channel Attention Module (BCAM) is designed, assigning greater weights to channels corresponding to critical information. Next, a lightweight Decoupled Head (LD-Head) is devised. Additionally, the loss function is redesigned to address asynchrony between labels and confidences, alleviating the imbalance between positive and negative samples during the neural network training. Finally, an adaptive scale factor for Intersection over Union (IoU) calculation is innovatively proposed, adjusting bounding box sizes adaptively to accommodate targets of different sizes in the dataset. Experimental results on the SODA10M and CBIA8K datasets demonstrate that ESF-YOLO increases Average Precision at 0.50 IoU (AP50) by 3.93 and 2.24%, Average Precision at 0.75 IoU (AP75) by 4.77 and 4.85%, and mean Average Precision (mAP) by 4 and 5.39%, respectively, validating the model's broad applicability.

A Serial Multi-Scale Feature Fusion and Enhancement Network for Amur Tiger Re-Identification

The Amur tiger is an important endangered species in the world, and its re-identification (re-ID) plays an important role in regional biodiversity assessment and wildlife resource statistics. This paper focuses on the task of Amur tiger re-ID based on visible light images from screenshots of surveillance videos or camera traps, aiming to solve the problem of low accuracy caused by camera perspective, noisy background noise, changes in motion posture, and deformation of Amur tiger body patterns during the re-ID process. To overcome this challenge, we propose a serial multi-scale feature fusion and enhancement re-ID network of Amur tiger for this task, in which global and local branches are constructed. Specifically, we design a global inverted pyramid multi-scale feature fusion method in the global branch to effectively fuse multi-scale global features and achieve high-level, fine-grained, and deep semantic feature preservation. We also design a local dual-domain attention feature enhancement method in the local branch, further enhancing local feature extraction and fusion by dividing local feature blocks. Based on the above model structure, we evaluated the effectiveness and feasibility of the model on the public dataset of the Amur Tiger Re-identification in the Wild (ATRW), and achieved good results on mAP, Rank-1, and Rank-5, demonstrating a certain competitiveness. In addition, since our proposed model does not require the introduction of additional expensive annotation information and does not incorporate other pre-training modules, it has important advantages such as strong transferability and simple training.

2.5D UNet with context-aware feature sequence fusion for accurate esophageal tumor semantic segmentation

Segmenting esophageal tumor from computed tomography (CT) sequence images can assist doctors in diagnosing and treating patients with this malignancy. However, accurately extracting esophageal tumor features from CT images often present challenges due to their small area, variable position, and shape, as well as the low contrast with surrounding tissues. This results in not achieving the level of accuracy required for practical applications in current methods. To address this problem, we propose a 2.5D context-aware feature sequence fusion UNet (2.5D CFSF-UNet) model for esophageal tumor segmentation in CT sequence images. Specifically, we embed intra-slice multiscale attention feature fusion (Intra-slice MAFF) in each skip connection of UNet to improve feature learning capabilities, better expressing the differences between anatomical structures within CT sequence images. Additionally, the inter-slice context fusion block (Inter-slice CFB) is utilized in the center bridge of UNet to enhance the depiction of context features between CT slices, thereby preventing the loss of structural information between slices. Experiments are conducted on a dataset of 430 esophageal tumor patients. The results show an 87.13% dice similarity coefficient, a 79.71% intersection over union and a 2.4758 mm Hausdorff distance, which demonstrates that our approach can improve contouring consistency and can be applied to clinical applications.

Combining data augmentation and deep learning for improved epilepsy detection

Introduction

In recent years, the use of EEG signals for seizure detection has gained widespread academic attention. Aiming at the problem of overfitting deep learning models due to the small number of EEG signal data during epilepsy detection, this paper proposes an epilepsy detection method that combines data augmentation and deep learning.

Methods

First, the Adversarial and Mixup Data Augmentation (AMDA) method is used to realize the data augmentation, which effectively enriches the number of training samples. To further improve the classification accuracy and robustness of epilepsy detection, this paper proposes a one-dimensional convolutional neural network and gated recurrent unit (AM-1D CNN-GRU) network model based on attention mechanism for epilepsy detection.

Results and discussion

The experimental results show that the performance of epilepsy detection achieved by using augmented data is significantly improved, and the accuracy, sensitivity, and area under the subject's working characteristic curve are up to 96.06, 95.48%, and 0.9637, respectively. Compared with the non-augmented data, all indicators are increased by more than 6.2%. Meanwhile, the detection performance was significantly improved compared with other epilepsy detection methods. The results of this research can provide a reference for the clinical application of epilepsy detection.

MA-PEP: A novel anticancer peptide prediction framework with multimodal feature fusion based on attention mechanism

AntiCancer Peptides (ACPs) have emerged as promising therapeutic agents for cancer treatment. The time-consuming and costly nature of wet-lab discriminatory methods has spurred the development of various machine learning and deep learning-based ACP classification methods. Nonetheless, current methods encountered challenges in efficiently integrating features from various peptide modalities, thereby limiting a more comprehensive understanding of ACPs and further restricting the improvement of prediction model performance. In this study, we introduce a novel ACP prediction method, MA-PEP, which leverages multiple attention mechanisms for feature enhancement and fusion to improve ACP prediction. By integrating the enhanced molecular-level chemical features and sequence information of peptides, MA-PEP demonstrates superior prediction performance across several benchmark datasets, highlighting its efficacy in ACP prediction. Moreover, the visual analysis and case studies further demonstrate MA-PEP's reliable feature extraction capability and its promise in the realm of ACP exploration. The code and datasets for MA-PEP are available at https://github.com/liangxiaodata/MA-PEP.

Enhanced Multitask Learning for Hash Code Generation of Palmprint Biometrics

This paper presents a novel multitask learning framework for palmprint biometrics, which optimizes classification and hashing branches jointly. The classification branch within our framework facilitates the concurrent execution of three distinct tasks: identity recognition and classification of soft biometrics, encompassing gender and chirality. On the other hand, the hashing branch enables the generation of palmprint hash codes, optimizing for minimal storage as templates and efficient matching. The hashing branch derives the complementary information from these tasks by amalgamating knowledge acquired from the classification branch. This approach leads to superior overall performance compared to individual tasks in isolation. To enhance the effectiveness of multitask learning, two additional modules, an attention mechanism module and a customized gate control module, are introduced. These modules are vital in allocating higher weights to crucial channels and facilitating task-specific expert knowledge integration. Furthermore, an automatic weight adjustment module is incorporated to optimize the learning process further. This module fine-tunes the weights assigned to different tasks, improving performance. Integrating the three modules above has shown promising accuracies across various classification tasks and has notably improved authentication accuracy. The extensive experimental results validate the efficacy of our proposed framework.

Detecting tea tree pests in complex backgrounds using a hybrid architecture guided by transformers and multi-scale attention mechanism

BACKGROUND

Tea pests pose a significant threat to tea leaf yield and quality, necessitating fast and accurate detection methods to improve pest control efficiency and reduce economic losses for tea farmers. However, in real tea gardens, some tea pests are small in size and easily camouflaged by complex backgrounds, making it challenging for farmers to promptly and accurately identify them.

RESULTS

To address this issue, we propose a real-time detection method based on TP-YOLOX for monitoring tea pests in complex backgrounds. Our approach incorporates the CSBLayer module, which combines convolution and multi-head self-attention mechanisms, to capture global contextual information from images and expand the network's perception field. Additionally, we integrate an efficient multi-scale attention module to enhance the model's ability to perceive fine details in small targets. To expedite model convergence and improve the precision of target localization, we employ the SIOU loss function as the bounding box regression function. Experimental results demonstrate that TP-YOLOX achieves a significant performance improvement with a relatively small additional computational cost (0.98 floating-point operations), resulting in a 4.50% increase in mean average precision (mAP) compared to the original YOLOX-s. When compared with existing object detection algorithms, TP-YOLOX outperforms them in terms of mAP performance. Moreover, the proposed method achieves a frame rate of 82.66 frames per second, meeting real-time requirements.

CONCLUSION

TP-YOLOX emerges as a proficient solution, capable of accurately and swiftly identifying tea pests amidst the complex backgrounds of tea gardens. This contribution not only offers valuable insights for tea pest monitoring but also serves as a reference for achieving precise pest control. © 2023 Society of Chemical Industry.

Identification of circRNA-disease associations via multi-model fusion and ensemble learning

Circular RNA (circRNA) is a common non-coding RNA and plays an important role in the diagnosis and therapy of human diseases, circRNA-disease associations prediction based on computational methods can provide a new way for better clinical diagnosis. In this article, we proposed a novel method for circRNA-disease associations prediction based on ensemble learning, named ELCDA. First, the association heterogeneous network was constructed via collecting multiple information of circRNAs and diseases, and multiple similarity measures are adopted here, then, we use metapath, matrix factorization and GraphSAGE-based models to extract features of nodes from different views, the final comprehensive features of circRNAs and diseases via ensemble learning, finally, a soft voting ensemble strategy is used to integrate the predicted results of all classifier. The performance of ELCDA is evaluated by fivefold cross-validation and compare with other state-of-the-art methods, the experimental results show that ELCDA is outperformance than others. Furthermore, three common diseases are used as case studies, which also demonstrate that ELCDA is an effective method for predicting circRNA-disease associations.

Multi-Granularity Aggregation with Spatiotemporal Consistency for Video-Based Person Re-Identification

Video-based person re-identification (ReID) aims to exploit relevant features from spatial and temporal knowledge. Widely used methods include the part- and attention-based approaches for suppressing irrelevant spatial-temporal features. However, it is still challenging to overcome inconsistencies across video frames due to occlusion and imperfect detection. These mismatches make temporal processing ineffective and create an imbalance of crucial spatial information. To address these problems, we propose the Spatiotemporal Multi-Granularity Aggregation (ST-MGA) method, which is specifically designed to accumulate relevant features with spatiotemporally consistent cues. The proposed framework consists of three main stages: extraction, which extracts spatiotemporally consistent partial information; augmentation, which augments the partial information with different granularity levels; and aggregation, which effectively aggregates the augmented spatiotemporal information. We first introduce the consistent part-attention (CPA) module, which extracts spatiotemporally consistent and well-aligned attentive parts. Sub-parts derived from CPA provide temporally consistent semantic information, solving misalignment problems in videos due to occlusion or inaccurate detection, and maximize the efficiency of aggregation through uniform partial information. To enhance the diversity of spatial and temporal cues, we introduce the Multi-Attention Part Augmentation (MA-PA) block, which incorporates fine parts at various granular levels, and the Long-/Short-term Temporal Augmentation (LS-TA) block, designed to capture both long- and short-term temporal relations. Using densely separated part cues, ST-MGA fully exploits and aggregates the spatiotemporal multi-granular patterns by comparing relations between parts and scales. In the experiments, the proposed ST-MGA renders state-of-the-art performance on several video-based ReID benchmarks (i.e., MARS, DukeMTMC-VideoReID, and LS-VID).

SEB-YOLO: An Improved YOLOv5 Model for Remote Sensing Small Target Detection

Due to the limited semantic information extraction with small objects and difficulty in distinguishing similar targets, it brings great challenges to target detection in remote sensing scenarios, which results in poor detection performance. This paper proposes an improved YOLOv5 remote sensing image target detection algorithm, SEB-YOLO (SPD-Conv + ECSPP + Bi-FPN + YOLOv5). Firstly, the space-to-depth (SPD) layer followed by a non-strided convolution (Conv) layer module (SPD-Conv) was used to reconstruct the backbone network, which retained the global features and reduced the feature loss. Meanwhile, the pooling module with the attention mechanism of the final layer of the backbone network was designed to help the network better identify and locate the target. Furthermore, a bidirectional feature pyramid network (Bi-FPN) with bilinear interpolation upsampling was added to improve bidirectional cross-scale connection and weighted feature fusion. Finally, the decoupled head is introduced to enhance the model convergence and solve the contradiction between the classification task and the regression task. Experimental results on NWPU VHR-10 and RSOD datasets show that the mAP of the proposed algorithm reaches 93.5% and 93.9%respectively, which is 4.0% and 5.3% higher than that of the original YOLOv5l algorithm. The proposed algorithm achieves better detection results for complex remote sensing images.

Automatic Identification of Pangolin Behavior Using Deep Learning Based on Temporal Relative Attention Mechanism

With declining populations in the wild, captive rescue and breeding have become one of the most important ways to protect pangolins from extinction. At present, the success rate of artificial breeding is low, due to the insufficient understanding of the breeding behavior characteristics of pangolins. The automatic recognition method based on machine vision not only monitors for 24 h but also reduces the stress response of pangolins. This paper aimed to establish a temporal relation and attention mechanism network (Pangolin breeding attention and transfer network, PBATn) to monitor and recognize pangolin behaviors, including breeding and daily behavior. There were 11,476 videos including breeding behavior and daily behavior that were divided into training, validation, and test sets. For the training set and validation set, the PBATn network model had an accuracy of 98.95% and 96.11%, and a loss function value of 0.1531 and 0.1852. The model is suitable for a 2.40 m × 2.20 m (length × width) pangolin cage area, with a nest box measuring 40 cm × 30 cm × 30 cm (length × width × height) positioned either on the left or right side inside the cage. A spherical night-vision monitoring camera was installed on the cage wall at a height of 2.50 m above the ground. For the test set, the mean Average Precision (mAP), average accuracy, average recall, average specificity, and average F1 score were found to be higher than SlowFast, X3D, TANet, TSN, etc., with values of 97.50%, 99.17%, 97.55%, 99.53%, and 97.48%, respectively. The recognition accuracies of PBATn were 94.00% and 98.50% for the chasing and mounting breeding behaviors, respectively. The results showed that PBATn outperformed the baseline methods in all aspects. This study shows that the deep learning system can accurately observe pangolin breeding behavior and it will be useful for analyzing the behavior of these animals.

Lightweight semantic segmentation network for tumor cell nuclei and skin lesion

In the field of medical image segmentation, achieving fast and accurate semantic segmentation of tumor cell nuclei and skin lesions is of significant importance. However, the considerable variations in skin lesion forms and cell types pose challenges to attaining high network accuracy and robustness. Additionally, as network depth increases, the growing parameter size and computational complexity make practical implementation difficult. To address these issues, this paper proposes MD-UNet, a fast cell nucleus segmentation network that integrates Tokenized Multi-Layer Perceptron modules, attention mechanisms, and Inception structures. Firstly, tokenized MLP modules are employed to label and project convolutional features, reducing computational complexity. Secondly, the paper introduces Depthwise Attention blocks and Multi-layer Feature Extraction modules. The Depthwise Attention blocks eliminate irrelevant and noisy responses from coarse-scale extracted information, serving as alternatives to skip connections in the UNet architecture. The Multi-layer Feature Extraction modules capture a wider range of high-level and low-level semantic features during decoding and facilitate feature fusion. The proposed MD-UNet approach is evaluated on two datasets: the International Skin Imaging Collaboration (ISIC2018) dataset and the PanNuke dataset. The experimental results demonstrate that MD-UNet achieves the best performance on both datasets.

PUTransGCN: identification of piRNA-disease associations based on attention encoding graph convolutional network and positive unlabelled learning

Piwi-interacting RNAs (piRNAs) play a crucial role in various biological processes and are implicated in disease. Consequently, there is an escalating demand for computational tools to predict piRNA-disease interactions. Although there have been computational methods proposed for the detection of piRNA-disease associations, the problem of imbalanced and sparse dataset has brought great challenges to capture the complex relationships between piRNAs and diseases. In response to this necessity, we have developed a novel computational architecture, denoted as PUTransGCN, which uses heterogeneous graph convolutional networks to uncover potential piRNA-disease associations. Additionally, the attention mechanism was used to adjust the weight parameters of aggregation heterogeneous node features automatically. For tackling the imbalanced dataset problem, the combined positive unlabelled learning (PUL) method comprising PU bagging, two-step and spy technique was applied to select reliable negative associations. The features of piRNAs and diseases were derived from three distinct biological sources by PUTransGCN, including information on piRNA sequences, semantic terms related to diseases and the existing network of piRNA-disease associations. In the experiment, PUTransGCN performs in 5-fold cross-validation with an AUC of 0.93 and 0.95 on two datasets, respectively, which outperforms the other six state-of-the-art models. We compared three different PUL methods, and the results of the ablation experiment indicate that the combined PUL method yields the best results. The PUTransGCN could serve as a valuable piRNA-disease prediction tool for upcoming studies in the biomedical field. The code for PUTransGCN is available at https://github.com/chenqiuhao/PUTransGCN.

A fluorescence detection method for postharvest tomato epidermal defects based on improved YOLOv5m

BACKGROUND

Tomato quality visual grading is greatly affected by the problems of smooth skin, uneven illumination and invisible defects that are difficult to identify. The realization of intelligent detection of postharvest epidermal defects is conducive to further improving the economic value of postharvest tomatoes.

RESULTS

An image acquisition device that utilizes fluorescence technology has been designed to capture a dataset of tomato skin defects, encompassing categories such as rot defects, crack defects and imperceptible defects. The YOLOv5m model was improved by introducing Convolutional Block Attention Module and replacing part of the convolution kernels in the backbone network with Switchable Atrous Convolution. The results of comparison experiments and ablation experiments show that the Precision, Recall and mean Average Precision of the improved YOLOv5m model were 89.93%, 82.33% and 87.57%, which are higher than YOLOv5m, Faster R-CNN and YOLOv7, and the average detection time was reduced by 47.04 ms picture-1.

CONCLUSION

The present study utilizes fluorescence imaging and an improved YOLOv5m model to detect tomato epidermal defects, resulting in better identification of imperceptible defects and detection of multiple categories of defects. This provides strong technical support for intelligent detection and quality grading of tomatoes. © 2024 Society of Chemical Industry.

Goal-Guided Graph Attention Network with Interactive State Refinement for Multi-Agent Trajectory Prediction

The accurate prediction of the future trajectories of traffic participants is crucial for enhancing the safety and decision-making capabilities of autonomous vehicles. Modeling social interactions among agents and revealing the inherent relationships is crucial for accurate trajectory prediction. In this context, we propose a goal-guided and interaction-aware state refinement graph attention network (SRGAT) for multi-agent trajectory prediction. This model effectively integrates high-precision map data and dynamic traffic states and captures long-term temporal dependencies through the Transformer network. Based on these dependencies, it generates multiple potential goals and Points of Interest (POIs). Through its dual-branch, multimodal prediction approach, the model not only proposes various plausible future trajectories associated with these POIs, but also rigorously assesses the confidence levels of each trajectory. This goal-oriented strategy enables SRGAT to accurately predict the future movement trajectories of other vehicles in complex traffic scenarios. Tested on the Argoverse and nuScenes datasets, SRGAT surpasses existing algorithms in key performance metrics by adeptly integrating past trajectories and current context. This goal-guided approach not only enhances long-term prediction accuracy, but also ensures its reliability, demonstrating a significant advancement in trajectory forecasting.

Fully Automated Identification of Lymph Node Metastases and Lymphovascular Invasion in Endometrial Cancer From Multi-Parametric MRI by Deep Learning

BACKGROUND

Early and accurate identification of lymphatic node metastasis (LNM) and lymphatic vascular space invasion (LVSI) for endometrial cancer (EC) patients is important for treatment design, but difficult on multi-parametric MRI (mpMRI) images.

PURPOSE

To develop a deep learning (DL) model to simultaneously identify of LNM and LVSI of EC from mpMRI images.

STUDY TYPE

Retrospective.

POPULATION

Six hundred twenty-one patients with histologically proven EC from two institutions, including 111 LNM-positive and 168 LVSI-positive, divided into training, internal, and external test cohorts of 398, 169, and 54 patients, respectively.

FIELD STRENGTH/SEQUENCE

T2-weighted imaging (T2WI), contrast-enhanced T1WI (CE-T1WI), and diffusion-weighted imaging (DWI) were scanned with turbo spin-echo, gradient-echo, and two-dimensional echo-planar sequences, using either a 1.5 T or 3 T system.

ASSESSMENT

EC lesions were manually delineated on T2WI by two radiologists and used to train an nnU-Net model for automatic segmentation. A multi-task DL model was developed to simultaneously identify LNM and LVSI positive status using the segmented EC lesion regions and T2WI, CE-T1WI, and DWI images as inputs. The performance of the model for LNM-positive diagnosis was compared with those of three radiologists in the external test cohort.

STATISTICAL TESTS

Dice similarity coefficient (DSC) was used to evaluate segmentation results. Receiver Operating Characteristic (ROC) analysis was used to assess the performance of LNM and LVSI status identification. P value <0.05 was considered significant.

RESULTS

EC lesion segmentation model achieved mean DSC values of 0.700 ± 0.25 and 0.693 ± 0.21 in the internal and external test cohorts, respectively. For LNM positive/LVSI positive identification, the proposed model achieved AUC values of 0.895/0.848, 0.806/0.795, and 0.804/0.728 in the training, internal, and external test cohorts, respectively, and better than those of three radiologists (AUC = 0.770/0.648/0.674).

DATA CONCLUSION

The proposed model has potential to help clinicians to identify LNM and LVSI status of EC patients and improve treatment planning.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Wheat Seed Detection and Counting Method Based on Improved YOLOv8 Model

Wheat seed detection has important applications in calculating thousand-grain weight and crop breeding. In order to solve the problems of seed accumulation, adhesion, and occlusion that can lead to low counting accuracy, while ensuring fast detection speed with high accuracy, a wheat seed counting method is proposed to provide technical support for the development of the embedded platform of the seed counter. This study proposes a lightweight real-time wheat seed detection model, YOLOv8-HD, based on YOLOv8. Firstly, we introduce the concept of shared convolutional layers to improve the YOLOv8 detection head, reducing the number of parameters and achieving a lightweight design to improve runtime speed. Secondly, we incorporate the Vision Transformer with a Deformable Attention mechanism into the C2f module of the backbone network to enhance the network's feature extraction capability and improve detection accuracy. The results show that in the stacked scenes with impurities (severe seed adhesion), the YOLOv8-HD model achieves an average detection accuracy (mAP) of 77.6%, which is 9.1% higher than YOLOv8. In all scenes, the YOLOv8-HD model achieves an average detection accuracy (mAP) of 99.3%, which is 16.8% higher than YOLOv8. The memory size of the YOLOv8-HD model is 6.35 MB, approximately 4/5 of YOLOv8. The GFLOPs of YOLOv8-HD decrease by 16%. The inference time of YOLOv8-HD is 2.86 ms (on GPU), which is lower than YOLOv8. Finally, we conducted numerous experiments and the results showed that YOLOv8-HD outperforms other mainstream networks in terms of mAP, speed, and model size. Therefore, our YOLOv8-HD can efficiently detect wheat seeds in various scenarios, providing technical support for the development of seed counting instruments.

Synthesis of gadolinium-enhanced glioma images on multisequence magnetic resonance images using contrastive learning

BACKGROUND

Gadolinium-based contrast agents are commonly used in brain magnetic resonance imaging (MRI), however, they cannot be used by patients with allergic reactions or poor renal function. For long-term follow-up patients, gadolinium deposition in the body can cause nephrogenic systemic fibrosis and other potential risks.

PURPOSE

Developing a new method of enhanced image synthesis based on the advantages of multisequence MRI has important clinical value for these patients. In this paper, an end-to-end synthesis model structure similarity index measure (SSIM)-based Dual Constrastive Learning with Attention (SDACL) based on contrastive learning is proposed to synthesize contrast-enhanced T1 (T1ce) using three unenhanced MRI images of T1, T2, and Flair in patients with glioma.

METHODS

The model uses the attention-dilation generator to enlarge the receptive field by expanding the residual blocks and to strengthen the feature representation and context learning of multisequence MRI. To enhance the detail and texture performance of the imaged tumor area, a comprehensive loss function combining patch-level contrast loss and structural similarity loss is created, which can effectively suppress noise and ensure the consistency of synthesized images and real images.

RESULTS

The normalized root-mean-square error (NRMSE), peak signal-to-noise ratio (PSNR), and SSIM of the model on the independent test set are 0.307  ± $\pm$  0.12, 23.337  ± $\pm$  3.21, and 0.881  ± $\pm$  0.05, respectively.

CONCLUSIONS

Results show this method can be used for the multisequence synthesis of T1ce images, which can provide valuable information for clinical diagnosis.

A Metric-Based Few-Shot Learning Method for Fish Species Identification with Limited Samples

Fish species identification plays a vital role in marine fisheries resource exploration, yet datasets related to marine fish resources are scarce. In open-water environments, various fish species often exhibit similar appearances and sizes. To solve these issues, we propose a few-shot learning approach to identifying fish species. Our approach involves two key components. Firstly, the embedding module was designed to address the challenges posed by a large number of fish species with similar phenotypes by utilizing the distribution relationships of species in the embedding space. Secondly, a metric function was introduced, effectively enhancing the performance of fish species classification and successfully addressing the issue of limited sample quantity. The proposed model is trained end to end on fish species public datasets including the Croatian fish dataset, Fish4Knowledge and WildFish. Compared with the prototypical networks, our method performs more effectively and improves accuracy by 2% to 10%; it is able to identify fish effectively in small samples sizes and complex scene scenarios. This method provides a valuable technological tool for the development of fisheries resources and the preservation of fish biodiversity.

Advancing Ocular Imaging: A Hybrid Attention Mechanism-Based U-Net Model for Precise Segmentation of Sub-Retinal Layers in OCT Images

This paper presents a novel U-Net model incorporating a hybrid attention mechanism for automating the segmentation of sub-retinal layers in Optical Coherence Tomography (OCT) images. OCT is an ophthalmology tool that provides detailed insights into retinal structures. Manual segmentation of these layers is time-consuming and subjective, calling for automated solutions. Our proposed model combines edge and spatial attention mechanisms with the U-Net architecture to improve segmentation accuracy. By leveraging attention mechanisms, the U-Net focuses selectively on image features. Extensive evaluations using datasets demonstrate that our model outperforms existing approaches, making it a valuable tool for medical professionals. The study also highlights the model's robustness through performance metrics such as an average Dice score of 94.99%, Adjusted Rand Index (ARI) of 97.00%, and Strength of Agreement (SOA) classifications like "Almost Perfect", "Excellent", and "Very Strong". This advanced predictive model shows promise in expediting processes and enhancing the precision of ocular imaging in real-world applications.

Attention-based speech feature transfer between speakers

In this study, we propose a simple yet effective method for incorporating the source speaker's characteristics in the target speaker's speech. This allows our model to generate the speech of the target speaker with the style of the source speaker. To achieve this, we focus on the attention model within the speech synthesis model, which learns various speaker features such as spectrogram, pitch, intensity, formant, pulse, and voice breaks. The model is trained separately using datasets specific to the source and target speakers. Subsequently, we replace the attention weights learned from the source speaker's dataset with the attention weights from the target speaker's model. Finally, by providing new input texts to the target model, we generate the speech of the target speaker with the styles of the source speaker. We validate the effectiveness of our model through similarity analysis utilizing five evaluation metrics and showcase real-world examples.

Drug-target Affinity Prediction by Molecule Secondary Structure Representation Network

INTRODUCTION

Identification of drug-target interactions (DTI) is a crucial step in drug development with high specificity and low toxicity. To accelerate the process, computer-aided DTI prediction algorithms have been used to screen compounds or targets rapidly. Furthermore, DTI prediction can be used to identify potential targets for existing drugs, thus uncovering new indications and repositioning them. Therefore, it is of great importance to develop efficient and accurate DTI prediction algorithms.

METHOD

Current algorithms usually represent drugs as extracted features, which are learned by convolutional neural networks (CNNs) from its linear representation, or utilize graph neural networks (GNNs) to learn its graph representation. However, these methods either lose information or fail to capture the structural information of the drug. To address this issue, a novel molecule secondary structure representation network (MSSRN) is proposed to learn drug characterization more accurately. Firstly, the network performs relational graph convolutional networks (R-GCNs) on the drug's molecular graph and integrates drug sequence convolutions to learn the sequential information. Secondly, inspired by the attention mechanism, spatial importance weights of the drug sequence are calculated to guide R-GCNs to learn the topological information of the drug.

RESULT

A drug-target affinity model, called MSSRN-DTA, was then constructed by using MSSRN to learn drug structure and CNN to learn protein sequence.

CONCLUSION

The effectiveness of the proposed method is verified by comparing it with other alternative methods and baseline models on two benchmark datasets.

End-to-end model-based trajectory prediction for ro-ro ship route using dual-attention mechanism

With the rapid increase of economic globalization, the significant expansion of shipping volume has resulted in shipping route congestion, causing the necessity of trajectory prediction for effective service and efficient management. While trajectory prediction can achieve a relatively high level of accuracy, the performance and generalization of prediction models remain critical bottlenecks. Therefore, this article proposes a dual-attention (DA) based end-to-end (E2E) neural network (DAE2ENet) for trajectory prediction. In the E2E structure, long short-term memory (LSTM) units are included for the task of pursuing sequential trajectory data from the encoder layer to the decoder layer. In DA mechanisms, global attention is introduced between the encoder and decoder layers to facilitate interactions between input and output trajectory sequences, and multi-head self-attention is utilized to extract sequential features from the input trajectory. In experiments, we use a ro-ro ship with a fixed navigation route as a case study. Compared with baseline models and benchmark neural networks, DAE2ENet can obtain higher performance on trajectory prediction, and better validation of environmental factors on ship navigation.

Developing a New Constitutive Model of High Damping Rubber by Combining GRU and Attention Mechanism

High damping rubber (HDR) bearings are extensively used in seismic design for bridges due to their remarkable energy dissipation capabilities, which is critical during earthquakes. A thorough assessment of crucial factors such as temperature, rate, experienced maximum amplitude, and the Mullins effect of HDR on the mechanics-based constitutive model of HDR is lacking. To address this issue, we propose a deep learning approach that integrates the Gate Recurrent Unit (GRU) and attention mechanism to identify time series characteristics from compression-shear test data of HDR specimens. It is shown that the combination of GRU and attention mechanism enables accurate prediction of the mechanical behavior of HDR specimens. Compared to the sole use of GRU, this suggested method significantly reduces model complexity and computation time while maintaining good prediction performance. Therefore, it offers a new approach to constructing the HDR constitutive model. Finally, the HDR constitutive model was used to analyze the impact of experienced maximum amplitudes and cycles on following processes. It was observed that maximum amplitudes directly influence the stress-strain relationship of HDR during subsequent processes. Consequently, a solid foundation is laid for evaluating the responses of HDR bearings under earthquakes.

Classifying breast cancer using multi-view graph neural network based on multi-omics data

Introduction: As the evaluation indices, cancer grading and subtyping have diverse clinical, pathological, and molecular characteristics with prognostic and therapeutic implications. Although researchers have begun to study cancer differentiation and subtype prediction, most of relevant methods are based on traditional machine learning and rely on single omics data. It is necessary to explore a deep learning algorithm that integrates multi-omics data to achieve classification prediction of cancer differentiation and subtypes. Methods: This paper proposes a multi-omics data fusion algorithm based on a multi-view graph neural network (MVGNN) for predicting cancer differentiation and subtype classification. The model framework consists of a graph convolutional network (GCN) module for learning features from different omics data and an attention module for integrating multi-omics data. Three different types of omics data are used. For each type of omics data, feature selection is performed using methods such as the chi-square test and minimum redundancy maximum relevance (mRMR). Weighted patient similarity networks are constructed based on the selected omics features, and GCN is trained using omics features and corresponding similarity networks. Finally, an attention module integrates different types of omics features and performs the final cancer classification prediction. Results: To validate the cancer classification predictive performance of the MVGNN model, we conducted experimental comparisons with traditional machine learning models and currently popular methods based on integrating multi-omics data using 5-fold cross-validation. Additionally, we performed comparative experiments on cancer differentiation and its subtypes based on single omics data, two omics data, and three omics data. Discussion: This paper proposed the MVGNN model and it performed well in cancer classification prediction based on multiple omics data.

Time series prediction of insect pests in tea gardens

BACKGROUND

Tea-garden pest control is crucial to ensure tea quality. In this context, the time-series prediction of insect pests in tea gardens is very important. Deep-learning-based time-series prediction techniques are advancing rapidly but research into their use in tea-garden pest prediction is limited. The current study investigates the time-series prediction of whitefly populations in the Tea Expo Garden, Jurong City, Jiangsu Province, China, employing three deep-learning algorithms, namely Informer, the Long Short-Term Memory (LSTM) network, and LSTM-Attention.

RESULTS

The comparative analysis of the three deep-learning algorithms revealed optimal results for LSTM-Attention, with an average root mean square error (RMSE) of 2.84 and average mean absolute error (MAE) of 2.52 for 7 days' prediction length, respectively. For a prediction length of 3 days, LSTM achieved the best performance, with an average RMSE of 2.60 and an average MAE of 2.24.

CONCLUSION

These findings suggest that different prediction lengths influence model performance in tea garden pest time series prediction. Deep learning could be applied satisfactorily to predict time series of insect pests in tea gardens based on LSTM-Attention. Thus, this study provides a theoretical basis for the research on the time series of pest and disease infestations in tea plants. © 2024 Society of Chemical Industry.

A Light Multi-View Stereo Method with Patch-Uncertainty Awareness

Multi-view stereo methods utilize image sequences from different views to generate a 3D point cloud model of the scene. However, existing approaches often overlook coarse-stage features, impacting the final reconstruction accuracy. Moreover, using a fixed range for all the pixels during inverse depth sampling can adversely affect depth estimation. To address these challenges, we present a novel learning-based multi-view stereo method incorporating attention mechanisms and an adaptive depth sampling strategy. Firstly, we propose a lightweight, coarse-feature-enhanced feature pyramid network in the feature extraction stage, augmented by a coarse-feature-enhanced module. This module integrates features with channel and spatial attention, enriching the contextual features that are crucial for the initial depth estimation. Secondly, we introduce a novel patch-uncertainty-based depth sampling strategy for depth refinement, dynamically configuring depth sampling ranges within the GRU-based optimization process. Furthermore, we incorporate an edge detection operator to extract edge features from the reference image's feature map. These edge features are additionally integrated into the iterative cost volume construction, enhancing the reconstruction accuracy. Lastly, our method is rigorously evaluated on the DTU and Tanks and Temples benchmark datasets, revealing its low GPU memory consumption and competitive reconstruction quality compared to other learning-based MVS methods.

Advancing Transcription Factor Binding Site Prediction Using DNA Breathing Dynamics and Sequence Transformers via Cross Attention

Understanding the impact of genomic variants on transcription factor binding and gene regulation remains a key area of research, with implications for unraveling the complex mechanisms underlying various functional effects. Our study delves into the role of DNA's biophysical properties, including thermodynamic stability, shape, and flexibility in transcription factor (TF) binding. We developed a multi-modal deep learning model integrating these properties with DNA sequence data. Trained on ChIP-Seq (chromatin immunoprecipitation sequencing) data in vivo involving 690 TF-DNA binding events in human genome, our model significantly improves prediction performance in over 660 binding events, with up to 9.6% increase in AUROC metric compared to the baseline model when using no DNA biophysical properties explicitly. Further, we expanded our analysis to in vitro high-throughput Systematic Evolution of Ligands by Exponential enrichment (SELEX) and Protein Binding Microarray (PBM) datasets, comparing our model with established frameworks. The inclusion of DNA breathing features consistently improved TF binding predictions across different cell lines in these datasets. Notably, for complex ChIP-Seq datasets, integrating DNABERT2 with a cross-attention mechanism provided greater predictive capabilities and insights into the mechanisms of disease-related non-coding variants found in genome-wide association studies. This work highlights the importance of DNA biophysical characteristics in TF binding and the effectiveness of multi-modal deep learning models in gene regulation studies.

A Deep Convolutional Neural Network for Pneumonia Detection in X-ray Images with Attention Ensemble

In the domain of AI-driven healthcare, deep learning models have markedly advanced pneumonia diagnosis through X-ray image analysis, thus indicating a significant stride in the efficacy of medical decision systems. This paper presents a novel approach utilizing a deep convolutional neural network that effectively amalgamates the strengths of EfficientNetB0 and DenseNet121, and it is enhanced by a suite of attention mechanisms for refined pneumonia image classification. Leveraging pre-trained models, our network employs multi-head, self-attention modules for meticulous feature extraction from X-ray images. The model's integration and processing efficiency are further augmented by a channel-attention-based feature fusion strategy, one that is complemented by a residual block and an attention-augmented feature enhancement and dynamic pooling strategy. Our used dataset, which comprises a comprehensive collection of chest X-ray images, represents both healthy individuals and those affected by pneumonia, and it serves as the foundation for this research. This study delves deep into the algorithms, architectural details, and operational intricacies of the proposed model. The empirical outcomes of our model are noteworthy, with an exceptional performance marked by an accuracy of 95.19%, a precision of 98.38%, a recall of 93.84%, an F1 score of 96.06%, a specificity of 97.43%, and an AUC of 0.9564 on the test dataset. These results not only affirm the model's high diagnostic accuracy, but also highlight its promising potential for real-world clinical deployment.

PathoGraph: An Attention-Based Graph Neural Network Capable of Prognostication Based on CD276 Labelling of Malignant Glioma Cells

Computerized methods have been developed that allow quantitative morphological analyses of whole slide images (WSIs), e.g., of immunohistochemical stains. The latter are attractive because they can provide high-resolution data on the distribution of proteins in tissue. However, many immunohistochemical results are complex because the protein of interest occurs in multiple locations (in different cells and also extracellularly). We have recently established an artificial intelligence framework, PathoFusion which utilises a bifocal convolutional neural network (BCNN) model for detecting and counting arbitrarily definable morphological structures. We have now complemented this model by adding an attention-based graph neural network (abGCN) for the advanced analysis and automated interpretation of such data. Classical convolutional neural network (CNN) models suffer from limitations when handling global information. In contrast, our abGCN is capable of creating a graph representation of cellular detail from entire WSIs. This abGCN method combines attention learning with visualisation techniques that pinpoint the location of informative cells and highlight cell-cell interactions. We have analysed cellular labelling for CD276, a protein of great interest in cancer immunology and a potential marker of malignant glioma cells/putative glioma stem cells (GSCs). We are especially interested in the relationship between CD276 expression and prognosis. The graphs permit predicting individual patient survival on the basis of GSC community features. Our experiments lay a foundation for the use of the BCNN-abGCN tool chain in automated diagnostic prognostication using immunohistochemically labelled histological slides, but the method is essentially generic and potentially a widely usable tool in medical research and AI based healthcare applications.

A Contraband Detection Scheme in X-ray Security Images Based on Improved YOLOv8s Network Model

X-ray inspections of contraband are widely used to maintain public transportation safety and protect life and property when people travel. To improve detection accuracy and reduce the probability of missed and false detection, a contraband detection algorithm YOLOv8s-DCN-EMA-IPIO* based on YOLOv8s is proposed. Firstly, the super-resolution reconstruction method based on the SRGAN network enhances the original data set, which is more conducive to model training. Secondly, DCNv2 (deformable convolution net v2) is introduced in the backbone network and merged with the C2f layer to improve the ability of the feature extraction and robustness of the model. Then, an EMA (efficient multi-scale attention) mechanism is proposed to suppress the interference of complex background noise and occlusion overlap in the detection process. Finally, the IPIO (improved pigeon-inspired optimization), which is based on the cross-mutation strategy, is employed to maximize the convolutional neural network's learning rate to derive the optimal group's weight information and ultimately improve the model's detection and recognition accuracy. The experimental results show that on the self-built data set, the mAP (mean average precision) of the improved model YOLOv8s-DCN-EMA-IPIO* is 73.43%, 3.98% higher than that of the original model YOLOv8s, and the FPS is 95, meeting the deployment requirements of both high precision and real-time.

Colorectal image analysis for polyp diagnosis

Colorectal polyp is an important early manifestation of colorectal cancer, which is significant for the prevention of colorectal cancer. Despite timely detection and manual intervention of colorectal polyps can reduce their chances of becoming cancerous, most existing methods ignore the uncertainties and location problems of polyps, causing a degradation in detection performance. To address these problems, in this paper, we propose a novel colorectal image analysis method for polyp diagnosis via PAM-Net. Specifically, a parallel attention module is designed to enhance the analysis of colorectal polyp images for improving the certainties of polyps. In addition, our method introduces the GWD loss to enhance the accuracy of polyp diagnosis from the perspective of polyp location. Extensive experimental results demonstrate the effectiveness of the proposed method compared with the SOTA baselines. This study enhances the performance of polyp detection accuracy and contributes to polyp detection in clinical medicine.

Finger Vein Identification Based on Large Kernel Convolution and Attention Mechanism

FV (finger vein) identification is a biometric identification technology that extracts the features of FV images for identity authentication. To address the limitations of CNN-based FV identification, particularly the challenge of small receptive fields and difficulty in capturing long-range dependencies, an FV identification method named Let-Net (large kernel and attention mechanism network) was introduced, which combines local and global information. Firstly, Let-Net employs large kernels to capture a broader spectrum of spatial contextual information, utilizing deep convolution in conjunction with residual connections to curtail the volume of model parameters. Subsequently, an integrated attention mechanism is applied to augment information flow within the channel and spatial dimensions, effectively modeling global information for the extraction of crucial FV features. The experimental results on nine public datasets show that Let-Net has excellent identification performance, and the EER and accuracy rate on the FV_USM dataset can reach 0.04% and 99.77%. The parameter number and FLOPs of Let-Net are only 0.89M and 0.25G, which means that the time cost of training and reasoning of the model is low, and it is easier to deploy and integrate into various applications.

DeepPPThermo: A Deep Learning Framework for Predicting Protein Thermostability Combining Protein-Level and Amino Acid-Level Features

Using wet experimental methods to discover new thermophilic proteins or improve protein thermostability is time-consuming and expensive. Machine learning methods have shown powerful performance in the study of protein thermostability in recent years. However, how to make full use of multiview sequence information to predict thermostability effectively is still a challenge. In this study, we proposed a deep learning-based classifier named DeepPPThermo that fuses features of classical sequence features and deep learning representation features for classifying thermophilic and mesophilic proteins. In this model, deep neural network (DNN) and bi-long short-term memory (Bi-LSTM) are used to mine hidden features. Furthermore, local attention and global attention mechanisms give different importance to multiview features. The fused features are fed to a fully connected network classifier to distinguish thermophilic and mesophilic proteins. Our model is comprehensively compared with advanced machine learning algorithms and deep learning algorithms, proving that our model performs better. We further compare the effects of removing different features on the classification results, demonstrating the importance of each feature and the robustness of the model. Our DeepPPThermo model can be further used to explore protein diversity, identify new thermophilic proteins, and guide directed mutations of mesophilic proteins.

CNN and Attention-Based Joint Source Channel Coding for Semantic Communications in WSNs

Wireless Sensor Networks (WSNs) have emerged as an efficient solution for numerous real-time applications, attributable to their compactness, cost-effectiveness, and ease of deployment. The rapid advancement of 5G technology and mobile edge computing (MEC) in recent years has catalyzed the transition towards large-scale deployment of WSN devices. However, the resulting data proliferation and the dynamics of communication environments introduce new challenges for WSN communication: (1) ensuring robust communication in adverse environments and (2) effectively alleviating bandwidth pressure from massive data transmission. In response to the aforementioned challenges, this paper proposes a semantic communication solution. Specifically, considering the limited computational and storage resources of WSN devices, we propose a flexible Attention-based Adaptive Coding (AAC) module. This module integrates window and channel attention mechanisms, dynamically adjusts semantic information in response to the current channel state, and facilitates adaptation of a single model across various Signal-to-Noise Ratio (SNR) environments. Furthermore, to validate the effectiveness of this approach, the paper introduces an end-to-end Joint Source Channel Coding (JSCC) scheme for image semantic communication, employing the AAC module. Experimental results demonstrate that the proposed scheme surpasses existing deep JSCC schemes across datasets of varying resolutions; furthermore, they validate the efficacy of the proposed AAC module, which is capable of dynamically adjusting critical information according to the current channel state. This enables the model to be trained over a range of SNRs and obtain better results.

Dynamic Serpentine Convolution with Attention Mechanism Enhancement for Beef Cattle Behavior Recognition

Behavior recognition in beef cattle is a crucial component of beef cattle behavior warning and intelligent farming. Traditional beef cattle behavior recognition faces challenges in both difficulty in identification and low accuracy. In this study, the YOLOv8n_BiF_DSC (Fusion of Dynamic Snake Convolution and BiFormer Attention) algorithm was employed for the non-intrusive recognition of beef cattle behavior. The specific steps are as follows: 45 beef cattle were observed using a fixed camera (A LINE OF DEFENSE) and a mobile phone (Huawei Mate20Pro) to collect and filter posture data, yielding usable videos ranging from 1 to 30 min in length. These videos cover nine different behaviors in various scenarios, including standing, lying, mounting, fighting, licking, eating, drinking, walking, and searching. After data augmentation, the dataset comprised 34,560 samples. The convolutional layer (CONV) was improved by introducing variable convolution and dynamic snake-like convolution modules. The dynamic snake-like convolution, which yielded the best results, expanded the model's receptive field, dynamically perceived key features of beef cattle behavior, and enhanced the algorithm's feature extraction capability. Attention mechanism modules, including SE (Squeeze-and-Excitation Networks), CBAM (Convolutional Block Attention Module), CA (Coordinate Attention), and BiFormer (Vision Transformer with Bi-Level Routing Attention), were introduced. The BiFormer attention mechanism, selected for its optimal performance, improved the algorithm's ability to capture long-distance context dependencies. The model's computational efficiency was enhanced through dynamic and query-aware perception. Experimental results indicated that YOLOv8n_BiF_DSC achieved the best results among all improved algorithms in terms of accuracy, average precision at IoU 50, and average precision at IoU 50:95. The accuracy of beef cattle behavior recognition reached 93.6%, with the average precision at IoU 50 and IoU 50:95 being 96.5% and 71.5%, respectively. This represents a 5.3%, 5.2%, and 7.1% improvement over the original YOLOv8n. Notably, the average accuracy of recognizing the lying posture of beef cattle reached 98.9%. In conclusion, the YOLOv8n_BiF_DSC algorithm demonstrates excellent performance in feature extraction and high-level data fusion, displaying high robustness and adaptability. It provides theoretical and practical support for the intelligent recognition and management of beef cattle.

Fault Diagnosis Methods for an Artillery Loading System Driving Motor in Complex Noisy Environments

With the development of modern military technology, electrical drive technology has become a power source for modern artillery. In fault monitoring of a driving motor mounted on a piece of artillery, various sensors are susceptible to interference from the complex environment, both inside and outside the artillery itself. In this study, we creatively propose a fault diagnosis model based on an attention mechanism, the AdaBoost method and a wavelet noise reduction network to address the difficulty in obtaining high-quality motor signals in complex noisy interference environments. First, multiple fusion wavelet basis, soft thresholding, and index soft filter optimization were used to train multiple wavelet noise reduction networks that could recover sample signals under different noise conditions. Second, a convolutional neural network (CNN) classification module was added to construct end-to-end classification models that could correctly identify faults. The above basis classification models were then integrated into the AdaBoost method with an improved attention mechanism to develop a fault diagnosis model suitable for complex noisy environments. Finally, two experiments were conducted to validate the proposed method. Under motor signals with varying signal-to-noise ratios (SNRs) noises, the proposed method achieved an average accuracy of 92%, surpassing the conventional method by over 8.5%.

Attenphos: General Phosphorylation Site Prediction Model Based on Attention Mechanism

Phosphorylation site prediction has important application value in the field of bioinformatics. It can act as an important reference and help with protein function research, protein structure research, and drug discovery. So, it is of great significance to propose scientific and effective calculation methods to accurately predict phosphorylation sites. In this study, we propose a new method, Attenphos, based on the self-attention mechanism for predicting general phosphorylation sites in proteins. The method not only captures the long-range dependence information of proteins but also better represents the correlation between amino acids through feature vector encoding transformation. Attenphos takes advantage of the one-dimensional convolutional layer to reduce the number of model parameters, improve model efficiency and prediction accuracy, and enhance model generalization. Comparisons between our method and existing state-of-the-art prediction tools were made using balanced datasets from human proteins and unbalanced datasets from mouse proteins. We performed prediction comparisons using independent test sets. The results showed that Attenphos demonstrated the best overall performance in the prediction of Serine (S), Threonine (T), and Tyrosine (Y) sites on both balanced and unbalanced datasets. Compared to current state-of-the-art methods, Attenphos has significantly higher prediction accuracy. This proves the potential of Attenphos in accelerating the identification and functional analysis of protein phosphorylation sites and provides new tools and ideas for biological research and drug discovery.

A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis

Introduction

Multiple sclerosis (MS) is a chronic neurological disorder characterized by the progressive loss of myelin and axonal structures in the central nervous system. Accurate detection and monitoring of MS-related changes in brain structures are crucial for disease management and treatment evaluation. We propose a deep learning algorithm for creating Voxel-Guided Morphometry (VGM) maps from longitudinal MRI brain volumes for analyzing MS disease activity. Our approach focuses on developing a generalizable model that can effectively be applied to unseen datasets.

Methods

Longitudinal MS patient high-resolution 3D T1-weighted follow-up imaging from three different MRI systems were analyzed. We employed a 3D residual U-Net architecture with attention mechanisms. The U-Net serves as the backbone, enabling spatial feature extraction from MRI volumes. Attention mechanisms are integrated to enhance the model's ability to capture relevant information and highlight salient regions. Furthermore, we incorporate image normalization by histogram matching and resampling techniques to improve the networks' ability to generalize to unseen datasets from different MRI systems across imaging centers. This ensures robust performance across diverse data sources.

Results

Numerous experiments were conducted using a dataset of 71 longitudinal MRI brain volumes of MS patients. Our approach demonstrated a significant improvement of 4.3% in mean absolute error (MAE) against the state-of-the-art (SOTA) method. Furthermore, the algorithm's generalizability was evaluated on two unseen datasets (n = 116) with an average improvement of 4.2% in MAE over the SOTA approach.

Discussion

Results confirm that the proposed approach is fast and robust and has the potential for broader clinical applicability.

Aircraft Engine Fault Diagnosis Model Based on 1DCNN-BiLSTM with CBAM

As the operational status of aircraft engines evolves, their fault modes also undergo changes. In response to the operational degradation trend of aircraft engines, this paper proposes an aircraft engine fault diagnosis model based on 1DCNN-BiLSTM with CBAM. The model can be directly applied to raw monitoring data without the need for additional algorithms to extract fault degradation features. It fully leverages the advantages of 1DCNN in extracting local features along the spatial dimension and incorporates CBAM, a channel and spatial attention mechanism. CBAM could assign higher weights to features relevant to fault categories and make the model pay more attention to them. Subsequently, it utilizes BiLSTM to handle nonlinear time feature sequences and bidirectional contextual feature information. Finally, experimental validation is conducted on the publicly available CMAPSS dataset from NASA, categorizing fault modes into three types: faultless, HPC fault (the single fault), and HPC&Fan fault (the mixed fault). Comparative analysis with other models reveals that the proposed model has a higher classification accuracy, which is of practical significance in improving the reliability of aircraft engine operations and for Remaining Useful Life (RUL) prediction.

Wearable Sensor-Based Residual Multifeature Fusion Shrinkage Networks for Human Activity Recognition

Human activity recognition (HAR) based on wearable sensors has emerged as a low-cost key-enabling technology for applications such as human-computer interaction and healthcare. In wearable sensor-based HAR, deep learning is desired for extracting human active features. Due to the spatiotemporal dynamic of human activity, a special deep learning network for recognizing the temporal continuous activities of humans is required to improve the recognition accuracy for supporting advanced HAR applications. To this end, a residual multifeature fusion shrinkage network (RMFSN) is proposed. The RMFSN is an improved residual network which consists of a multi-branch framework, a channel attention shrinkage block (CASB), and a classifier network. The special multi-branch framework utilizes a 1D-CNN, a lightweight temporal attention mechanism, and a multi-scale feature extraction method to capture diverse activity features via multiple branches. The CASB is proposed to automatically select key features from the diverse features for each activity, and the classifier network outputs the final recognition results. Experimental results have shown that the accuracy of the proposed RMFSN for the public datasets UCI-HAR, WISDM, and OPPORTUNITY are 98.13%, 98.35%, and 93.89%, respectively. In comparison with existing advanced methods, the proposed RMFSN could achieve higher accuracy while requiring fewer model parameters.

CL-SPO2Net: Contrastive Learning Spatiotemporal Attention Network for Non-Contact Video-Based SpO2 Estimation

Video-based peripheral oxygen saturation (SpO2) estimation, utilizing solely RGB cameras, offers a non-contact approach to measuring blood oxygen levels. Previous studies set a stable and unchanging environment as the premise for non-contact blood oxygen estimation. Additionally, they utilized a small amount of labeled data for system training and learning. However, it is challenging to train optimal model parameters with a small dataset. The accuracy of blood oxygen detection is easily affected by ambient light and subject movement. To address these issues, this paper proposes a contrastive learning spatiotemporal attention network (CL-SPO2Net), an innovative semi-supervised network for video-based SpO2 estimation. Spatiotemporal similarities in remote photoplethysmography (rPPG) signals were found in video segments containing facial or hand regions. Subsequently, integrating deep neural networks with machine learning expertise enabled the estimation of SpO2. The method had good feasibility in the case of small-scale labeled datasets, with the mean absolute error between the camera and the reference pulse oximeter of 0.85% in the stable environment, 1.13% with lighting fluctuations, and 1.20% in the facial rotation situation.

Recognition of rare antinuclear antibody patterns based on a novel attention-based enhancement framework

Rare antinuclear antibody (ANA) pattern recognition has been a widely applied technology for routine ANA screening in clinical laboratories. In recent years, the application of deep learning methods in recognizing ANA patterns has witnessed remarkable advancements. However, the majority of studies in this field have primarily focused on the classification of the most common ANA patterns, while another subset has concentrated on the detection of mitotic metaphase cells. To date, no prior research has been specifically dedicated to the identification of rare ANA patterns. In the present paper, we introduce a novel attention-based enhancement framework, which was designed for the recognition of rare ANA patterns in ANA-indirect immunofluorescence images. More specifically, we selected the algorithm with the best performance as our target detection network by conducting comparative experiments. We then further developed and enhanced the chosen algorithm through a series of optimizations. Then, attention mechanism was introduced to facilitate neural networks in expediting the learning process, extracting more essential and distinctive features for the target features that belong to the specific patterns. The proposed approach has helped to obtained high precision rate of 86.40%, 82.75% recall, 84.24% F1 score and 84.64% mean average precision for a 9-category rare ANA pattern detection task on our dataset. Finally, we evaluated the potential of the model as medical technologist assistant and observed that the technologist's performance improved after referring to the results of the model prediction. These promising results highlighted its potential as an efficient and reliable tool to assist medical technologists in their clinical practice.

scAMAC: self-supervised clustering of scRNA-seq data based on adaptive multi-scale autoencoder

Cluster assignment is vital to analyzing single-cell RNA sequencing (scRNA-seq) data to understand high-level biological processes. Deep learning-based clustering methods have recently been widely used in scRNA-seq data analysis. However, existing deep models often overlook the interconnections and interactions among network layers, leading to the loss of structural information within the network layers. Herein, we develop a new self-supervised clustering method based on an adaptive multi-scale autoencoder, called scAMAC. The self-supervised clustering network utilizes the Multi-Scale Attention mechanism to fuse the feature information from the encoder, hidden and decoder layers of the multi-scale autoencoder, which enables the exploration of cellular correlations within the same scale and captures deep features across different scales. The self-supervised clustering network calculates the membership matrix using the fused latent features and optimizes the clustering network based on the membership matrix. scAMAC employs an adaptive feedback mechanism to supervise the parameter updates of the multi-scale autoencoder, obtaining a more effective representation of cell features. scAMAC not only enables cell clustering but also performs data reconstruction through the decoding layer. Through extensive experiments, we demonstrate that scAMAC is superior to several advanced clustering and imputation methods in both data clustering and reconstruction. In addition, scAMAC is beneficial for downstream analysis, such as cell trajectory inference. Our scAMAC model codes are freely available at https://github.com/yancy2024/scAMAC.

Automatic detection of standing dead trees based on improved YOLOv7 from airborne remote sensing imagery

Detecting and localizing standing dead trees (SDTs) is crucial for effective forest management and conservation. Due to challenges posed by mountainous terrain and road conditions, conducting a swift and comprehensive survey of SDTs through traditional manual inventory methods is considerably difficult. In recent years, advancements in deep learning and remote sensing technology have facilitated real-time and efficient detection of dead trees. Nevertheless, challenges persist in identifying individual dead trees in airborne remote sensing images, attributed to factors such as small target size, mutual occlusion and complex backgrounds. These aspects collectively contribute to the increased difficulty of detecting dead trees at a single-tree scale. To address this issue, the paper introduces an improved You Only Look Once version 7 (YOLOv7) model that incorporates the Simple Parameter-Free Attention Module (SimAM), an unparameterized attention mechanism. This improvement aims to enhance the network's feature extraction capabilities and increase the model's sensitivity to small target dead trees. To validate the superiority of SimAM_YOLOv7, we compared it with four widely adopted attention mechanisms. Additionally, a method to enhance model robustness is presented, involving the replacement of the Complete Intersection over Union (CIoU) loss in the original YOLOv7 model with the Wise-IoU (WIoU) loss function. Following these, we evaluated detection accuracy using a self-developed dataset of SDTs in forests. The results indicate that the improved YOLOv7 model can effectively identify dead trees in airborne remote sensing images, achieving precision, recall and mAP@0.5 values of 94.31%, 93.13% and 98.03%, respectively. These values are 3.67%, 2.28% and 1.56% higher than those of the original YOLOv7 model. This improvement model provides a convenient solution for forest management.

ADH-Enhancer: an attention-based deep hybrid framework for enhancer identification and strength prediction

Enhancers play an important role in the process of gene expression regulation. In DNA sequence abundance or absence of enhancers and irregularities in the strength of enhancers affects gene expression process that leads to the initiation and propagation of diverse types of genetic diseases such as hemophilia, bladder cancer, diabetes and congenital disorders. Enhancer identification and strength prediction through experimental approaches is expensive, time-consuming and error-prone. To accelerate and expedite the research related to enhancers identification and strength prediction, around 19 computational frameworks have been proposed. These frameworks used machine and deep learning methods that take raw DNA sequences and predict enhancer's presence and strength. However, these frameworks still lack in performance and are not useful in real time analysis. This paper presents a novel deep learning framework that uses language modeling strategies for transforming DNA sequences into statistical feature space. It applies transfer learning by training a language model in an unsupervised fashion by predicting a group of nucleotides also known as k-mers based on the context of existing k-mers in a sequence. At the classification stage, it presents a novel classifier that reaps the benefits of two different architectures: convolutional neural network and attention mechanism. The proposed framework is evaluated over the enhancer identification benchmark dataset where it outperforms the existing best-performing framework by 5%, and 9% in terms of accuracy and MCC. Similarly, when evaluated over the enhancer strength prediction benchmark dataset, it outperforms the existing best-performing framework by 4%, and 7% in terms of accuracy and MCC.

CenterNet-Saccade: Enhancing Sonar Object Detection with Lightweight Global Feature Extraction

Sonar imaging technology is widely used in the field of marine and underwater monitoring because sound waves can be transmitted in elastic media, such as the atmosphere and seawater, without much interference. In underwater object detection, due to the unique characteristics of the monitored sonar image, and since the target in an image is often accompanied by its own shadow, we can use the relative relationship between the shadow and the target for detection. To make use of shadow-information-aided detection and realize accurate real-time detection in sonar images, we put forward a network based on a lightweight module. By using the attention mechanism with a global receptive field, the network can make the target pay attention to the shadow information in the global environment, and because of its exquisite design, the computational time of the network is greatly reduced. Specifically, we design a ShuffleBlock model adapted to Hourglass to make the backbone network lighter. The concept of CNN dimension reduction is applied to MHSA to make it more efficient while paying attention to global features. Finally, CenterNet's unreasonable distribution method of positive and negative samples is improved. Simulation experiments were carried out using the proposed sonar object detection dataset. The experimental results further verify that our improved model has obvious advantages over many existing conventional deep learning models. Moreover, the real-time monitoring performance of our proposed model is more conducive to the implementation in the field of ocean monitoring.