Emotion recognition of EEG signals based on contrastive learning graph convolutional model.

Objective.Electroencephalogram (EEG) signals offer invaluable insights into the complexities of emotion generation within the brain. Yet, the variability in EEG signals across individuals presents a formidable obstacle for empirical implementations. Our research addresses these challenges innovatively, focusing on the commonalities within distinct subjects' EEG data.Approach.We introduce a novel approach named Contrastive Learning Graph Convolutional Network (CLGCN). This method captures the distinctive features and crucial channel nodes related to individuals' emotional states. Specifically, CLGCN merges the dual benefits of CL's synchronous multisubject data learning and the GCN's proficiency in deciphering brain connectivity matrices. Understanding multifaceted brain functions and their information interchange processes is realized as CLGCN generates a standardized brain network learning matrix during a dataset's learning process.Main results.Our model underwent rigorous testing on the Database for Emotion Analysis using Physiological Signals (DEAP) and SEED datasets. In the five-fold cross-validation used for dependent subject experimental setting, it achieved an accuracy of 97.13% on the DEAP dataset and surpassed 99% on the SEED and SEED_IV datasets. In the incremental learning experiments with the SEED dataset, merely 5% of the data was sufficient to fine-tune the model, resulting in an accuracy of 92.8% for the new subject. These findings validate the model's efficacy.Significance.This work combines CL with GCN, improving the accuracy of decoding emotional states from EEG signals and offering valuable insights into uncovering the underlying mechanisms of emotional processes in the brain.

Pseudo-Spectral Spatial Feature Extraction and Enhanced Fusion Image for Efficient Meter-Sized Lunar Impact Crater Automatic Detection in Digital Orthophoto Map.

Impact craters are crucial for our understanding of planetary resources, geological ages, and the history of evolution. We designed a novel pseudo-spectral spatial feature extraction and enhanced fusion (PSEF) method with the YOLO network to address the problems encountered during the detection of the numerous and densely distributed meter-sized impact craters on the lunar surface. The illumination incidence edge features, isotropic edge features, and eigen frequency features are extracted by Sobel filtering, LoG filtering, and frequency domain bandpass filtering, respectively. Then, the PSEF images are created by pseudo-spectral spatial techniques to preserve additional details from the original DOM data. Moreover, we conducted experiments using the DES method to optimize the post-processing parameters of the models, thereby determining the parameter ranges for practical deployment. Compared with the Basal model, the PSEF model exhibited superior performance, as indicated by multiple measurement metrics, including the precision, recall, F1-score, mAP, and robustness, etc. Additionally, a statistical analysis of the error metrics of the predicted bounding boxes shows that the PSEF model performance is excellent in predicting the size, shape, and location of impact craters. These advancements offer a more accurate and consistent method to detect the meter-sized craters on planetary surfaces, providing crucial support for the exploration and study of celestial bodies in our solar system.

Doppler Radar Sensor-Based Fall Detection Using a Convolutional Bidirectional Long Short-Term Memory Model.

Falls among the elderly are a common and serious health risk that can lead to physical injuries and other complications. To promptly detect and respond to fall events, radar-based fall detection systems have gained widespread attention. In this paper, a deep learning model is proposed based on the frequency spectrum of radar signals, called the convolutional bidirectional long short-term memory (CB-LSTM) model. The introduction of the CB-LSTM model enables the fall detection system to capture both temporal sequential and spatial features simultaneously, thereby enhancing the accuracy and reliability of the detection. Extensive comparison experiments demonstrate that our model achieves an accuracy of 98.83% in detecting falls, surpassing other relevant methods currently available. In summary, this study provides effective technical support using the frequency spectrum and deep learning methods to monitor falls among the elderly through the design and experimental validation of a radar-based fall detection system, which has great potential for improving quality of life for the elderly and providing timely rescue measures.

One-Year-Old Precocious Chinese Mitten Crab Identification Algorithm Based on Task Alignment.

The cultivation of the Chinese mitten crab (Eriocheir sinensis) is an important component of China's aquaculture industry and also a field of concern worldwide. It focuses on the selection of high-quality, disease-free juvenile crabs. However, the early maturity rate of more than 18.2% and the mortality rate of more than 60% make it difficult to select suitable juveniles for adult culture. The juveniles exhibit subtle distinguishing features, and the methods for differentiating between sexes vary significantly; without training from professional breeders, it is challenging for laypersons to identify and select the appropriate juveniles. Therefore, we propose a task-aligned detection algorithm for identifying one-year-old precocious Chinese mitten crabs, named R-TNET. Initially, the required images were obtained by capturing key frames, and then they were annotated and preprocessed by professionals to build a training dataset. Subsequently, the ResNeXt network was selected as the backbone feature extraction network, with Convolutional Block Attention Modules (CBAMs) and a Deformable Convolution Network (DCN) embedded in its residual blocks to enhance its capability to extract complex features. Adaptive spatial feature fusion (ASFF) was then integrated into the feature fusion network to preserve the detailed features of small targets such as one-year-old precocious Chinese mitten crab juveniles. Finally, based on the detection head proposed by task-aligned one-stage object detection, the parameters of its anchor alignment metric were adjusted to detect, locate, and classify the crab juveniles. The experimental results showed that this method achieves a mean average precision (mAP) of 88.78% and an F1-score of 97.89%. This exceeded the best-performing mainstream object detection algorithm, YOLOv7, by 4.17% in mAP and 1.77% in the F1-score. Ultimately, in practical application scenarios, the algorithm effectively identified one-year-old precocious Chinese mitten crabs, providing technical support for the automated selection of high-quality crab juveniles in the cultivation process, thereby promoting the rapid development of aquaculture and agricultural intelligence in China.

A Unified Spatio-Temporal Inference Network for Car-Sharing Serial Prediction.

Car-sharing systems require accurate demand prediction to ensure efficient resource allocation and scheduling decisions. However, developing precise predictive models for vehicle demand remains a challenging problem due to the complex spatio-temporal relationships. This paper introduces USTIN, the Unified Spatio-Temporal Inference Prediction Network, a novel neural network architecture for demand prediction. The model consists of three key components: a temporal feature unit, a spatial feature unit, and a spatio-temporal feature unit. The temporal unit utilizes historical demand data and comprises four layers, each corresponding to a different time scale (hourly, daily, weekly, and monthly). Meanwhile, the spatial unit incorporates contextual points of interest data to capture geographic demand factors around parking stations. Additionally, the spatio-temporal unit incorporates weather data to model the meteorological impacts across locations and time. We conducted extensive experiments on real-world car-sharing data. The proposed USTIN model demonstrated its ability to effectively learn intricate temporal, spatial, and spatiotemporal relationships, and outperformed existing state-of-the-art approaches. Moreover, we employed negative binomial regression with uncertainty to identify the most influential factors affecting car usage.

A spatial feature analysis of primary health care utilization in a large city in China and its implications for family doctor contract service policy.

BACKGROUND: Family doctor contract policy is now run by the State Council as an important move to promote the hierarchical medical system. Whether the family doctor contract policy achieves the initial government's goal should be measured further from the perspective of patient visits between hospitals and community health centers, which are regarded as grass medical agencies. METHODS: The spatial feature measurement method is applied with ArcGIS 10.2 software to analyze the spatial aggregation effect of patient visits to hospitals or community health centers among 20 districts of one large city in China and analyze the family doctor contract policy published in those areas to compare the influence of visit tendencies. RESULTS: From year 2016-2020, visits to hospitals were in the high-high cluster, and the density was spatially overflow, while there was no such tendency in visits to community health centers. The analysis of different family doctor contract policy implementation times in 20 districts reflects that the family doctor contract policy has a very limited effect on the promotion of the hierarchical medical system, and the innovation of the family doctor contract policy needs to be considered. CONCLUSIONS: A brief summary and potential implications. A multi-integrated medical system along with family doctor contract policy needs to be established, especially integrated in leadership and governance, financing, workforce, and service delivery between hospitals and community health centers, to promote the hierarchical medical system.

Enhancing hazardous material vehicle detection with advanced feature enhancement modules using HMV-YOLO.

The transportation of hazardous chemicals on roadways has raised significant safety concerns. Incidents involving these substances often lead to severe and devastating consequences. Consequently, there is a pressing need for real-time detection systems tailored for hazardous material vehicles. However, existing detection methods face challenges in accurately identifying smaller targets and achieving high precision. This paper introduces a novel solution, HMV-YOLO, an enhancement of the YOLOv7-tiny model designed to address these challenges. Within this model, two innovative modules, CBSG and G-ELAN, are introduced. The CBSG module's mathematical model incorporates components such as Convolution (Conv2d), Batch Normalization (BN), SiLU activation, and Global Response Normalization (GRN) to mitigate feature collapse issues and enhance neuron activity. The G-ELAN module, building upon CBSG, further advances feature fusion. Experimental results showcase the superior performance of the enhanced model compared to the original one across various evaluation metrics. This advancement shows great promise for practical applications, particularly in the context of real-time monitoring systems for hazardous material vehicles.

Space-CNN: a decision classification method based on EEG signals from different brain regions.

Decision-making plays a critical role in an individual's interpersonal interactions and cognitive processes. Due to the issue of strong subjectivity in the classification research of art design decisions, we utilize the relatively objective electroencephalogram (EEG) to explore design decision problems. However, different regions of the brain do not have the same influence on the design decision classification, so this paper proposes a spatial feature based convolutional neural network (space-CNN) to explore the problem of decision classification of EEG signals from different regions. We recruit 16 subjects to collect their EEG data while viewing four stimulation patterns. After noise reduction of the raw data by discrete wavelet transform (DWT), the EEG image is generated by combining it with the spatial features of the EEG signal, which is used as the input of CNN. Our experimental results show that the degree of influence of different brain regions on decision-making is parietal lobe > frontal lobe > occipital lobe > temporal lobe. In addition, the average accuracy of space-CNN reaches 86.13%, which is about 6% higher than similar studies.

A novel feature enhancement and semantic segmentation scheme for identifying low-contrast ocean oil spills.

The oil spill accidents on the sea surface pose a severe threat to the marine environment and human health. This paper proposes a novel Semantic Segmentation Network (SSN) for processing oil spill images so that low-contrast oil spills on the sea surface can be accurately identified. After the detection accuracy and real-time performance of the current SSNs are compared, the basic network architecture of DeeplabV3+ based target detection is analyzed. The standard convolution is replaced by the Omni-dimensional Dynamic Convolution (ODConv) in the Ghost Module Depth-Wise separable Convolution (DWConv) to further enhance the feature extraction ability of the network. Furthermore, a new DeeplabV3+ based network with ODGhostNetV2 is constructed as the main feature extraction module, and an Adaptive Triplet Attention (ATA) module is deployed in the encoder and decoder at the same time. This not only improves the richness of semantic features but also increases the following receptive fields of the network model. ATA integrates the Adaptively Spatial Feature Fusion (ASFF) module to optimize the weight assignment problem in the feature map fusion process. The ablation experiments are conducted to verify the proposed network which show high accuracy and good real-time performance for the oil spill detection.

Robotic Grasp Detection Network Based on Improved Deformable Convolution and Spatial Feature Center Mechanism.

In this article, we propose an effective grasp detection network based on an improved deformable convolution and spatial feature center mechanism (DCSFC-Grasp) to precisely grasp unidentified objects. DCSFC-Grasp includes three key procedures as follows. First, improved deformable convolution is introduced to adaptively adjust receptive fields for multiscale feature information extraction. Then, an efficient spatial feature center (SFC) layer is explored to capture the global remote dependencies through a lightweight multilayer perceptron (MLP) architecture. Furthermore, a learnable feature center (LFC) mechanism is reported to gather local regional features and preserve the local corner region. Finally, a lightweight CARAFE operator is developed to upsample the features. Experimental results show that DCSFC-Grasp achieves a high accuracy (99.3% and 96.1% for the Cornell and Jacquard grasp datasets, respectively) and even outperforms the existing state-of-the-art grasp detection models. The results of real-world experiments on the six-DoF Realman RM65 robotic arm further demonstrate that our DCSFC-Grasp is effective and robust for the grasping of unknown targets.

An integrated network based on 2D/3D feature correlations for benign-malignant tumor classification and uncertainty estimation in digital breast tomosynthesis.

Objective.Classification of benign and malignant tumors is important for the early diagnosis of breast cancer. Over the last decade, digital breast tomosynthesis (DBT) has gradually become an effective imaging modality for breast cancer diagnosis due to its ability to generate three-dimensional (3D) visualizations. However, computer-aided diagnosis (CAD) systems based on 3D images require high computational costs and time. Furthermore, there is considerable redundant information in 3D images. Most CAD systems are designed based on 2D images, which may lose the spatial depth information of tumors. In this study, we propose a 2D/3D integrated network for the diagnosis of benign and malignant breast tumors.Approach.We introduce a correlation strategy to describe feature correlations between slices in 3D volumes, corresponding to the tissue relationship and spatial depth features of tumors. The correlation strategy can be used to extract spatial features with little computational cost. In the prediction stage, 3D spatial correlation features and 2D features are both used for classification.Main results.Experimental results demonstrate that our proposed framework achieves higher accuracy and reliability than pure 2D or 3D models. Our framework has a high area under the curve of 0.88 and accuracy of 0.82. The parameter size of the feature extractor in our framework is only 35% of that of the 3D models. In reliability evaluations, our proposed model is more reliable than pure 2D or 3D models because of its effective and nonredundant features.Significance.This study successfully combines 3D spatial correlation features and 2D features for the diagnosis of benign and malignant breast tumors in DBT. In addition to high accuracy and low computational cost, our model is more reliable and can output uncertainty value. From this point of view, the proposed method has the potential to be applied in clinic.

Early sensory gain control is dominated by obligatory and global feature-based attention in top-down shifts of combined spatial and feature-based attention.

What are the dynamics of global feature-based and spatial attention, when deployed together? In an attentional shifting experiment, flanked by three control experiments, we investigated neural temporal dynamics of combined attentional shifts. For this purpose, orange- and blue-frequency-tagged spatially overlapping Random Dot Kinematograms were presented in the left and right visual hemifield to elicit continuous steady-state-visual-evoked-potentials. After being initially engaged in a fixation cross task, participants were at some point in time cued to shift attention to one of the Random Dot Kinematograms, to detect and respond to brief coherent motion events, while ignoring all such events in other Random Dot Kinematograms. The analysis of steady-state visual-evoked potentials allowed us to map time courses and dynamics of early sensory-gain modulations by attention. This revealed a time-invariant amplification of the to-be attended color both at the attended and the unattended side, followed by suppression for the to-be-ignored color at attended and unattended sides. Across all experiments, global and obligatory feature-based selection dominated early sensory gain modulations, whereas spatial attention played a minor modulatory role. However, analyses of behavior and neural markers such as alpha-band activity and event-related potentials to target- and distractor-event processing, revealed clear modulations by spatial attention.

A novel residual block: replace Conv1× 1 with Conv3×3 and stack more convolutions.

The residual structure has an important influence on the design of the neural network model. The neural network model based on residual structure has excellent performance in computer vision tasks. However, the performance of classical residual networks is restricted by the size of receptive fields, channel information, spatial information and other factors. In this article, a novel residual structure is proposed. We modify the identity mapping and down-sampling block to get greater effective receptive field, and its excellent performance in channel information fusion and spatial feature extraction is verified by ablation studies. In order to further verify its feature extraction capability, a non-deep convolutional neural network (CNN) was designed and tested on Cifar10 and Cifar100 benchmark platforms using a naive training method. Our network model achieves better performance than other mainstream networks under the same training parameters, the accuracy we achieved is 3.08 percentage point higher than ResNet50 and 1.38 percentage points higher than ResNeXt50. Compared with SeResNet152, it is 0.29 percentage point higher in the case of 50 epochs less training.

Enhanced robust spatial feature selection and correlation filter learning for UAV tracking.

Spatial boundary effect can significantly reduce the performance of a learned discriminative correlation filter (DCF) model. A commonly used method to relieve this effect is to extract appearance features from a wider region of a target. However, this way would introduce unexpected features from background pixels and noises, which will lead to a decrease of the filter's discrimination power. To address this shortcoming, this paper proposes an innovative method called enhanced robust spatial feature selection and correlation filter Learning (EFSCF), which performs jointly sparse feature learning to handle boundary effects effectively while suppressing the influence of background pixels and noises. Unlike the ℓ2-norm-based tracking approaches that are prone to non-Gaussian noises, the proposed method imposes the ℓ2,1-norm on the loss term to enhance the robustness against the training outliers. To enhance the discrimination further, a jointly sparse feature selection scheme based on the ℓ2,1 -norm is designed to regularize the filter in rows and columns simultaneously. To the best of the authors' knowledge, this has been the first work exploring the structural sparsity in rows and columns of a learned filter simultaneously. The proposed model can be efficiently solved by an alternating direction multiplier method. The proposed EFSCF is verified by experiments on four challenging unmanned aerial vehicle datasets under severe noise and appearance changes, and the results show that the proposed method can achieve better tracking performance than the state-of-the-art trackers.

Reconstruction of Gap-Free Land Surface Temperature at a 100 m Spatial Resolution from Multidimensional Data: A Case in Wuhan, China.

Land surface temperatures (LST) are vital parameters in land surface-atmosphere interactions. Constrained by technology and atmospheric interferences, LST retrievals from various satellite sensors usually return missing data, thus negatively impacting analyses. Reconstructing missing data is important for acquiring gap-free datasets. However, the current reconstruction methods are limited for maintaining spatial details and high accuracies. We developed a new gap-free algorithm termed the spatial feature-considered random forest regression (SFRFR) model; it builds stable nonlinear relationships to connect the LST with related parameters, including terrain elements, land coverage types, spectral indexes, surface reflectance data, and the spatial feature of the LST, to reconstruct the missing LST data. The SFRFR model reconstructed gap-free LST data retrieved from the Landsat 8 satellite on 27 July 2017 in Wuhan. The results show that the SFRFR model exhibits the best performance according to the various evaluation metrics among the SFRFR, random forest regression and spline interpolation, with a coefficient of determination (R2) reaching 0.96, root-mean-square error (RMSE) of 0.55, and mean absolute error (MAE) of 0.55. Then, we reconstructed gap-free LST data gathered in Wuhan from 2016 to 2021 to analyze urban thermal environment changes and found that 2020 presented the coolest temperatures. The SFRFR model still displayed satisfactory results, with an average R2 of 0.91 and an MAE of 0.63. We further discuss and discover the factors affecting the visual performance of SFRFR and identify the research priority to circumvent these disadvantages. Overall, this study provides a simple, practical method for acquiring gap-free LST data to help us better understand the spatiotemporal LST variation process.

Breast Tumor Classification in Ultrasound Images by Fusion of Deep Convolutional Neural Network and Shallow LBP Feature.

Breast cancer is one of the most dangerous and common cancers in women which leads to a major research topic in medical science. To assist physicians in pre-screening for breast cancer to reduce unnecessary biopsies, breast ultrasound and computer-aided diagnosis (CAD) have been used to distinguish between benign and malignant tumors. In this study, we proposed a CAD system for tumor diagnosis using a multi-channel fusion method and feature extraction structure based on multi-feature fusion on breast ultrasound (BUS) images. In the pre-processing stage, the multi-channel fusion method completed the color conversion of the BUS image to make it contain richer information. In the feature extraction stage, the pre-trained ResNet50 network was selected as the basic network, and three levels of features were combined based on adaptive spatial feature fusion (ASFF), and finally, the shallow local binary pattern (LBP) texture features were fused. Support vector machine (SVM) was used for comparative analysis. A retrospective analysis was carried out, and 1615 breast tumor images (572 benign and 1043 malignant) confirmed by pathological examinations were collected. After data processing and augmentation, for an independent test set consisting of 874 breast ultrasound images (457 benign and 417 malignant), the accuracy, precision, recall, specificity, F1 score, and AUC of our method were 96.91%, 98.75%, 94.72%, 98.91%, 0.97, and 0.991, respectively. The results show that the integration of shallow LBP texture features and multi-level depth features can more effectively improve the comprehensive performance of breast tumor diagnosis, and has strong clinical application value. Compared with the past methods, our proposed method is expected to realize the automatic diagnosis of breast tumors and provide an auxiliary tool for radiologists to accurately diagnose breast diseases.

Spatial-Temporal Characteristics and Influence Factors of Carbon Emission from Livestock Industry in China.

Animal husbandry is an important source of carbon emissions. As a large country, China must measure the carbon emissions from animal husbandry to reveal the spatial and temporal characteristics and determine the influencing factors to realize low-carbon animal husbandry and carbon emission reduction. In this paper, the carbon emissions of the livestock industry in each province of China were calculated with the emission coefficient method, considering the temperature change factor. The spatial and temporal characteristics and influencing factors of livestock industry carbon emissions were analyzed using the kernel density model, the spatial autocorrelation model, and the Tobit model. The results indicated that: (1) From 2000 to 2020, carbon emissions from the livestock industry in China experienced four stages: rapid rise, rapid decline, slow rise, and fluctuating decline, with an overall downward trend. Carbon emissions in the eastern and central regions showed a downward trend, while carbon emissions in the western regions showed an upward trend. (2) In terms of time, the relative gap in carbon emissions among the provinces narrowed first and then widened; the spatial agglomeration of carbon emissions from livestock farming in China increased, gradually forming the characteristics of "high agglomeration, low agglomeration", and showing a gradually decreasing pattern from northwest to southeast. (3) Nationwide, industrial structure, population, and farmers' income levels have had significantly promoting effects on animal husbandry carbon emissions, and the urbanization and agricultural mechanization levels have had significant inhibitory effects on carbon emissions. Finally, based on the above factors, it can be concluded that recognizing the location conditions, promoting the upgrading of industrial structures, and adopting differentiated strategies will help to promote the reduction in carbon emissions in animal husbandry and achieve its high-quality development.

Scene Uyghur Text Detection Based on Fine-Grained Feature Representation.

Scene text detection task aims to precisely localize text in natural environments. At present, the application scenarios of text detection topics have gradually shifted from plain document text to more complex natural scenarios. Objects with similar texture and text morphology in the complex background noise of natural scene images are prone to false recall and difficult to detect multi-scale texts, a multi-directional scene Uyghur text detection model based on fine-grained feature representation and spatial feature fusion is proposed, and feature extraction and feature fusion are improved to enhance the network's ability to represent multi-scale features. In this method, the multiple groups of 3 × 3 convolutional feature groups that are connected like the hierarchical residual to build a residual network for feature extraction, which captures the feature details and increases the receptive field of the network to adapt to multi-scale text and long glued dimensional font detection and suppress false positives of text-like objects. Secondly, an adaptive multi-level feature map fusion strategy is adopted to overcome the inconsistency of information in multi-scale feature map fusion. The proposed model achieves 93.94% and 84.92% F-measure on the self-built Uyghur dataset and the ICDAR2015 dataset, respectively, which improves the accuracy of Uyghur text detection and suppresses false positives.

3D IFPN: Improved Feature Pyramid Network for Automatic Segmentation of Gastric Tumor.

Automatic segmentation of gastric tumor not only provides image-guided clinical diagnosis but also assists radiologists to read images and improve the diagnostic accuracy. However, due to the inhomogeneous intensity distribution of gastric tumors in CT scans, the ambiguous/missing boundaries, and the highly variable shapes of gastric tumors, it is quite challenging to develop an automatic solution. This study designs a novel 3D improved feature pyramidal network (3D IFPN) to automatically segment gastric tumors in computed tomography (CT) images. To meet the challenges of this extremely difficult task, the proposed 3D IFPN makes full use of the complementary information within the low and high layers of deep convolutional neural networks, which is equipped with three types of feature enhancement modules: 3D adaptive spatial feature fusion (ASFF) module, single-level feature refinement (SLFR) module, and multi-level feature refinement (MLFR) module. The 3D ASFF module adaptively suppresses the feature inconsistency in different levels and hence obtains the multi-level features with high feature invariance. Then, the SLFR module combines the adaptive features and previous multi-level features at each level to generate the multi-level refined features by skip connection and attention mechanism. The MLFR module adaptively recalibrates the channel-wise and spatial-wise responses by adding the attention operation, which improves the prediction capability of the network. Furthermore, a stage-wise deep supervision (SDS) mechanism and a hybrid loss function are also embedded to enhance the feature learning ability of the network. CT volumes dataset collected in three Chinese medical centers was used to evaluate the segmentation performance of the proposed 3D IFPN model. Experimental results indicate that our method outperforms state-of-the-art segmentation networks in gastric tumor segmentation. Moreover, to explore the generalization for other segmentation tasks, we also extend the proposed network to liver tumor segmentation in CT images of the MICCAI 2017 Liver Tumor Segmentation Challenge.

Breast ultrasound tumor image classification using image decomposition and fusion based on adaptive multi-model spatial feature fusion.

BACKGROUND AND OBJECTIVE: Breast cancer is a fatal threat to the health of women. Ultrasonography is a common method for the detection of breast cancer. Computer-aided diagnosis of breast ultrasound images can help doctors in diagnosing benign and malignant lesions. In this paper, by combining image decomposition and fusion techniques with adaptive spatial feature fusion technology, a reliable classification method for breast ultrasound images of tumors is proposed. METHODS: First, fuzzy enhancement and bilateral filtering algorithms are used to process the original breast ultrasound image. Then, various decomposition images representing the clinical characteristics of breast tumors are obtained using the original and mask images. By considering the diversity of the benign and malignant characteristic information represented by each decomposition image, the decomposition images are fused through the RGB channel, and three types of fusion images are generated. Then, from a series of candidate deep learning models, transfer learning is used to select the best model as the base model to extract deep learning features. Finally, while training the classification network, adaptive spatial feature fusion technology is used to train the weight network to complete deep learning feature fusion and classification. RESULTS: In this study, 1328 breast ultrasound images were collected for training and testing. The experimental results show that the values of accuracy, precision, specificity, sensitivity/recall, F1 score, and area under the curve of the proposed method were 0.9548, 0.9811, 0.9833, 0.9392, 0.9571, and 0.9883, respectively. CONCLUSION: Our research can automate breast cancer detection and has strong clinical utility. When compared to previous methods, our proposed method is expected to be more effective while assisting doctors in diagnosing breast ultrasound images.