Current studies on electroencephalogram (EEG) emotion recognition primarily concentrate on discrete stimulus paradigms under controlled laboratory settings, which cannot adequately represent the dynamic transition characteristics of emotional states during multi-context interactions. To address this issue, this paper proposes a novel method for emotion transition recognition that leverages a cross-modal feature fusion and global perception network (CFGPN). Firstly, an experimental paradigm encompassing six types of emotion transition scenarios was designed, and EEG and eye movement data were simultaneously collected from 20 participants, each annotated with dynamic continuous emotion labels. Subsequently, deep canonical correlation analysis integrated with a cross-modal attention mechanism was employed to fuse features from EEG and eye movement signals, resulting in multimodal feature vectors enriched with highly discriminative emotional information. These vectors are then input into a parallel hybrid architecture that combines convolutional neural networks (CNNs) and Transformers. The CNN is employed to capture local time-series features, whereas the Transformer leverages its robust global perception capabilities to effectively model long-range temporal dependencies, enabling accurate dynamic emotion transition recognition. The results demonstrate that the proposed method achieves the lowest mean square error in both valence and arousal recognition tasks on the dynamic emotion transition dataset and a classic multimodal emotion dataset. It exhibits superior recognition accuracy and stability when compared with five existing unimodal and six multimodal deep learning models. The approach enhances both adaptability and robustness in recognizing emotional state transitions in real-world scenarios, showing promising potential for applications in the field of biomedical engineering.
Current epilepsy prediction methods are not effective in characterizing the multi-domain features of complex long-term electroencephalogram (EEG) data, leading to suboptimal prediction performance. Therefore, this paper proposes a novel multi-scale sparse adaptive convolutional network based on multi-head attention mechanism (MS-SACN-MM) model to effectively characterize the multi-domain features. The model first preprocesses the EEG data, constructs multiple convolutional layers to effectively avoid information overload, and uses a multi-layer perceptron and multi-head attention mechanism to focus the network on critical pre-seizure features. Then, it adopts a focal loss training strategy to alleviate class imbalance and enhance the model's robustness. Experimental results show that on the publicly created dataset (CHB-MIT) by MIT and Boston Children's Hospital, the MS-SACN-MM model achieves a maximum accuracy of 0.999 for seizure prediction 10 ~ 15 minutes in advance. This demonstrates good predictive performance and holds significant importance for early intervention and intelligent clinical management of epilepsy patients.
In animal navigation, head direction is encoded by head direction cells within the olfactory-hippocampal structures of the brain. Even in darkness or unfamiliar environments, animals can estimate their head direction by integrating self-motion cues, though this process accumulates errors over time and undermines navigational accuracy. Traditional strategies rely on visual input to correct head direction, but visual scenes combined with self-motion information offer only partially accurate estimates. This study proposed an innovative calibration mechanism that dynamically adjusts the association between visual scenes and head direction based on the historical firing rates of head direction cells, without relying on specific landmarks. It also introduced a method to fine-tune error correction by modulating the strength of self-motion input to control the movement speed of the head direction cell activity bump. Experimental results showed that this approach effectively reduced the accumulation of self-motion-related errors and significantly enhanced the accuracy and robustness of the navigation system. These findings offer a new perspective for biologically inspired robotic navigation systems and underscore the potential of neural mechanisms in enabling efficient and reliable autonomous navigation.
With the rising incidence of breast cancer among women, neoadjuvant chemotherapy (NAC) is becoming increasingly crucial as a preoperative treatment modality, enabling tumor downstaging and volume reduction. However, its efficacy varies significantly among patients, underscoring the importance of predicting pathological complete response (pCR) following NAC. Early research relied on statistical methods to integrate clinical data for predicting treatment outcomes. With the advent of artificial intelligence (AI), traditional machine learning approaches were subsequently employed for efficacy prediction. Deep learning emerged to dominate this field, and demonstrated the capability to automatically extract imaging features and integrate multimodal data for pCR prediction. This review comprehensively examined the applications and limitations of these three methodologies in predicting breast cancer pCR. Future efforts must prioritize the development of superior predictive models to achieve precise predictions, integrate them into clinical workflows, enhance patient care, and ultimately improve therapeutic outcomes and quality of life.
Endometrial cancer (EC) is one of the most common gynecological malignancies, with an increasing incidence rate worldwide. Accurate segmentation of lesion areas in computed tomography (CT) images is a critical step in assisting clinical diagnosis. In this study, we propose a novel deep learning-based segmentation model, termed spatial choice and weight union network (SCWU-Net), which incorporates two newly designed modules: the spatial selection module (SSM) and the combination weight module (CWM). The SSM enhances the model's ability to capture contextual information through deep convolutional blocks, while the CWM, based on joint attention mechanisms, is employed within the skip connections to further boost segmentation performance. By integrating the strengths of both modules into a U-shaped multi-scale architecture, the model achieves precise segmentation of EC lesion regions. Experimental results on a public dataset demonstrate that SCWU-Net achieves a Dice similarity coefficient (DSC) of 82.98%, an intersection over union (IoU) of 78.63%, a precision of 92.36%, and a recall of 84.10%. Its overall performance is significantly outperforming other state-of-the-art models. This study enhances the accuracy of lesion segmentation in EC CT images and holds potential clinical value for the auxiliary diagnosis of endometrial cancer.
Stent migration is one of the common complications following transcatheter valve implantation. This study aims to design a "drum-shaped" balloon-expandable aortic valve stent to address this issue and conduct a mechanical analysis. The implantation process of the stent was evaluated using a method that combines numerical simulation and in vitro experiments. Furthermore, the fatigue process of the stent under pulsatile cyclic loading was simulated, and its fatigue performance was assessed using a Goodman diagram. The process of the stent migrating toward the left ventricular side was simulated, and the force-displacement curve of the stent was extracted to evaluate its anti- migration performance. The results showed that all five stent models could be crimped into a 14F sheath and enabled uniform expansion of the native valve leaflets. The stress in each stent was below the ultimate stress, so no fatigue fracture occurred. As the cell height ratio decreased, the contact area fraction between the stent and the aortic root gradually decreased. However, the mean contact force and the maximum anti-migration force first decreased and then increased. Specifically, model S5 had the smallest contact area fraction but the largest mean contact force and maximum anti-migration force, reaching approximately 0.16 MPa and 10.73 N, respectively. The designed stent achieves a "drum-shaped" change after expansion and has good anti-migration performance.
The three-dimensional (3D) printed bone tissue repair guide scaffold is considered a promising method for treating bone defect repair. In this experiment, chitosan (CS), sodium alginate (SA), and mineralized collagen (MC) were combined and 3D printed to form scaffolds. The experimental results showed that the printability of the scaffold was improved with the increase of chitosan concentration. Infrared spectroscopy analysis confirmed that the scaffold formed a cross-linked network through electrostatic interaction between chitosan and sodium alginate under acidic conditions, and X-ray diffraction results showed the presence of characteristic peaks of hydroxyapatite, indicating the incorporation of mineralized collagen into the scaffold system. In the in vitro collagen release experiments, a weakly alkaline environment was found to accelerate the release rate of collagen, and the release amount increased significantly with a lower concentration of chitosan. Cell experiments showed that scaffolds loaded with mineralized collagen could significantly promote cell proliferation activity and alkaline phosphatase expression. The subcutaneous implantation experiment further verified the biocompatibility of the material, and the implantation of printed scaffolds did not cause significant inflammatory reactions. Histological analysis showed no abnormal pathological changes in the surrounding tissues. Therefore, incorporating mineralized collagen into sodium alginate/chitosan scaffolds is believed to be a new tissue engineering and regeneration strategy for achieving enhanced osteogenic differentiation through the slow release of collagen.
It has been found that the incidence of cardiovascular disease in patients with lower limb amputation is significantly higher than that in normal people, and the risk of developing coronary atherosclerosis is much higher than that in other high-risk groups. Numerous studies have confirmed that high systolic and diastolic blood pressures are potential risk factors for coronary artery disease, and it has been demonstrated that the ascending aortic pressure during diastole increases after amputation. However, the relationship between lower limb amputation and coronary atherosclerosis has not been fully explained from the perspective of hemodynamic environment. Therefore, in this study, a centralized parameter model of the human cardiovascular system and a three-dimensional model of the left coronary artery were established to investigate the effect of amputation on the hemodynamic environment of the coronary artery. The results showed that the abnormal hemodynamic environment induced by amputation, characterized by factors such as increased diastolic pressure in the ascending aorta, led to a significant expansion of the low wall shear stress (WSS) region on the outer lateral aspect of the left coronary artery bifurcation during diastole. The maximum observed increase in the area of low WSS reached up to 50.5%. This abnormal hemodynamic environment elevates the risk of plaque formation in the left coronary artery. Moreover, the more severe the lower limb atrophy, the greater the risk of coronary atherosclerosis in amputees. This study preliminarily reveals the effect of lower limb amputation on the hemodynamic environment of the left coronary artery.
This study aims to investigate the therapeutic efficacy of 50 Hz-0.6 mT low-frequency pulsed electromagnetic field (PEMF) on postmenopausal osteoporosis in ovariectomized rats. Thirty 3-month-old female SD rats were selected and divided into a sham operation group (Sham), an ovariectomized model group (OVX), and a low-frequency pulsed electromagnetic field (PEMF) treatment group, with 10 rats in each group. After 8 weeks, the whole-body bone mineral density (BMD) of each group of rats was measured. The treatment group began to receive PEMF stimulation for 90 minutes daily, while the OVX group only received a simulated placement without electricity. After 6 weeks of intervention, all rats were sacrificed and tested for in vitro BMD, micro-CT, biomechanics, serum biochemical indicators, and bone tissue-related proteins. The results showed that the BMD of the OVX group was significantly lower than that of the Sham group 8 weeks after surgery, indicating successful modeling. After 6 weeks of treatment, compared with the OVX group, the PEMF group exhibited significantly increased BMD in the whole body, femur, and vertebral bodies. Micro-CT analysis results showed improved bone microstructure, significantly increased maximum load and bending strength of the femur, elevated levels of serum bone formation markers, and increased expression of osteogenic-related proteins. In conclusion, this study demonstrates that daily 90-minute exposure to 50 Hz-0.6 mT PEMF effectively enhances BMD, improves bone biomechanical properties, optimizes bone microstructure, stimulates bone formation, and inhibits bone resorption in ovariectomized rats, highlighting its therapeutic potential for postmenopausal osteoporosis.
Lung cancer is a leading cause of cancer-related deaths worldwide, with its high mortality rate primarily attributed to delayed diagnosis. Radiomics, by extracting abundant quantitative features from medical images, offers novel possibilities for early diagnosis and precise treatment of lung cancer. This article reviewed the latest advancements in radiomics for lung cancer management, particularly its integration with artificial intelligence (AI) to optimize diagnostic processes and personalize treatment strategies. Despite existing challenges, such as non-standardized image acquisition parameters and limitations in model reproducibility, the incorporation of AI significantly enhanced the precision and efficiency of image analysis, thereby improving the prediction of disease progression and the formulation of treatment plans. We emphasized the critical importance of standardizing image acquisition parameters and discussed the role of AI in advancing the clinical application of radiomics, alongside future research directions.
To facilitate the early intelligent screening of pediatric genu valgum, this study develops a deep learning-based gait recognition model tailored for clinical application. The model is constructed upon a three-dimensional residual network architecture and incorporates a triplet attention module alongside a spatial hierarchical pooling module, jointly enhancing feature interaction across temporal, spatial, and channel dimensions. This design ensures an optimal balance between representational capacity and computational efficiency. Evaluated on a self-constructed dataset, the model achieves precision of 98.0%, 97.1%, and 96.5%, recall rates of 97.5%, 97.0%, and 95.0%, and F 1-scores of 0.98, 0.97, and 0.96 on the training, validation, and test sets, respectively, demonstrating excellent recognition performance and strong generalization ability. Ablation experiments confirm the importance of the proposed model's core components in improving performance, and comparative experiments further highlight its significant advantages in recognition accuracy and robustness. Visualization experiments reveal that the model effectively focuses on key regions of gait images, with attention regions aligning closely with clinical anatomical landmarks, thereby enhancing the interpretability of the model's decision-making in clinical applications. In summary, the proposed model not only offers an efficient and reliable technical solution for early intelligent screening of genu valgum in children, but also provides a practical pathway for applying gait recognition technology in medical diagnosis.
In order to meet the need of autonomous control of patients with severe limb disorders, this paper designs a nursing bed control system based on motor imagery-brain computer interface (MI-BCI). In view of the low decoding performance of cross-subjects and the dynamic fluctuation of cognitive state in the existing MI-BCI technology, the neural network structure optimization and user interaction feedback enhancement are improved. Firstly, the optimized dual-branch graph convolution multi-scale neural network integrates dynamic graph convolution and multi-scale convolution. The average classification accuracy is higher than that of multi-scale attention temporal convolution network, Gram angle field combined with convolution long short term memory hybrid network, Transformer-based graph convolution network and other existing methods. Secondly, a dual visual feedback mechanism is constructed, in which electroencephalogram (EEG) topographic map feedback can improve the discrimination of spatial patterns, and attention state feedback can enhance the temporal stability of signals. Compared with the single EEG topographic map feedback and non-feedback system, the average classification accuracy of the proposed method is also greatly improved. Finally, in the four classification control task of nursing bed, the average control accuracy of the system is 90.84%, and the information transmission rate is 84.78 bits/min. In summary, this paper provides a reliable technical solution for improving the autonomous interaction ability of patients with severe limb disorders, which has important theoretical significance and application value.
Most current medical image segmentation models are primarily built upon the U-shaped network (U-Net) architecture, which has certain limitations in capturing both global contextual information and fine-grained details. To address this issue, this paper proposes a novel U-shaped network model, termed the Multi-View U-Net (MUNet), which integrates self-attention and multi-view attention mechanisms. Specifically, a newly designed multi-view attention module is introduced to aggregate semantic features from different perspectives, thereby enhancing the representation of fine details in images. Additionally, the MUNet model leverages a self-attention encoding block to extract global image features, and by fusing global and local features, it improves segmentation performance. Experimental results demonstrate that the proposed model achieves superior segmentation performance in coronary artery image segmentation tasks, significantly outperforming existing models. By incorporating self-attention and multi-view attention mechanisms, this study provides a novel and efficient modeling approach for medical image segmentation, contributing to the advancement of intelligent medical image analysis.
To address the challenges in blood cell recognition caused by diverse morphology, dense distribution, and the abundance of small target information, this paper proposes a blood cell detection algorithm - the "You Only Look Once" model based on hybrid mixing attention and deep over-parameters (HADO-YOLO). First, a hybrid attention mechanism is introduced into the backbone network to enhance the model's sensitivity to detailed features. Second, the standard convolution layers with downsampling in the neck network are replaced with deep over-parameterized convolutions to expand the receptive field and improve feature representation. Finally, the detection head is decoupled to enhance the model's robustness for detecting abnormal cells. Experimental results on the Blood Cell Counting Dataset (BCCD) demonstrate that the HADO-YOLO algorithm achieves a mean average precision of 90.2% and a precision of 93.8%, outperforming the baseline YOLO model. Compared with existing blood cell detection methods, the proposed algorithm achieves state-of-the-art detection performance. In conclusion, HADO-YOLO offers a more efficient and accurate solution for identifying various types of blood cells, providing valuable technical support for future clinical diagnostic applications.
Colorectal polyps are important early markers of colorectal cancer,and their early detection is crucial for cancer prevention.Although existing polyp segmentation models have achieved certain results,they still face challenges such as diverse polyp morphology,blurred boundaries,and insufficient feature extraction.To address these issues,this study proposes a parallel coordinate fusion network(PCFNet),aiming to improve the accuracy and robustness of polyp segmentation.PCFNet integrates parallel convolutional modules and a coordinate attention mechanism,enabling the preservation of global feature information while precisely capturing detailed features,thereby effectively segmenting polyps with complex boundaries.Experimental results on Kvasir-SEG and CVC-ClinicDB demonstrate the outstanding performance of PCFNet across multiple metrics.Specifically,on the Kvasir-SEG dataset,PCFNet achieved an F1-score of 0.897 4 and a mean intersection over union(mIoU)of 0.835 8;on the CVC-ClinicDB dataset,it attained an F1-score of 0.939 8 and an mIoU of 0.892 3.Compared with other methods,PCFNet shows significant improvements across all performance metrics,particularly in multi-scale feature fusion and spatial information capture,demonstrating its innovativeness.The proposed method provides a more reliable AI-assisted diagnostic tool for early colorectal cancer screening.
Subtype classification of age-related macular degeneration (AMD) based on optical coherence tomography (OCT) images serves as an effective auxiliary tool for clinicians in diagnosing disease progression and formulating treatment plans. To improve the classification accuracy of AMD subtypes, this study proposes a keypoint-based, attention-enhanced residual network (KPA-ResNet). The proposed architecture adopts a 50-layer residual network (ResNet-50) as the backbone, preceded by a keypoint localization module based on heatmap regression to outline critical lesion regions. A two-dimensional relative self-attention mechanism is incorporated into convolutional layers to enhance the representation of key lesion areas. Furthermore, the network depth is appropriately increased and an improved residual module, ConvNeXt, is introduced to enable comprehensive extraction of high-dimensional features and enrich the detail of lesion boundary contours, ultimately achieving higher classification accuracy of AMD subtypes. Experimental results demonstrate that KPA-ResNet achieves significant improvements in overall classification accuracy compared with conventional convolutional neural networks. Specifically, for the wet AMD subtypes, the classification accuracies for inactive choroidal neovascularization (CNV) and active CNV reach 92.8% and 95.2%, respectively, representing substantial improvement over ResNet-50. These findings validate the superior performance of KPA-ResNet in AMD subtype classification tasks. This work provides a high-accuracy, generalizable network architecture for OCT-based AMD subtype classification and offers new insights into integrating attention mechanisms with convolutional neural networks in ophthalmic image analysis.
The aim of this study is to investigate the protective effect of idebenone (IDE) combined with borneol (BO) against Parkinson's disease (PD). In this study, wild-type AB zebrafish and transgenic Tg ( vmat2: GFP) zebrafish with green fluorescence labeled dopamine neurons were used to establish the PD model with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine hydrochloride (MPTP). Following drug treatment, the behavioral performance and dopamine neuron morphology of zebrafish were evaluated, and regulation of dopamine signaling pathway-related genes was determined using RT-qPCR. The results showed that IDE combined with BO improved the behavioral disorders of zebrafish such as bradykinesia and shortening movement distance, also effectively reversed the damage of MPTP-induced dopaminergic neurons. At the same time, the expression of dopamine synthesis and transportation-related genes was up-regulated, and the normal function of the signal transduction pathway was restored. The combination showed a better therapeutic effect compared to the IDE monotherapy group. This study reveals the protective mechanism of IDE combined with BO on the central nervous system for the first time, which provides an important experimental basis and theoretical reference for clinical combination strategy in PD treatment.
Diabetic retinopathy is a common blinding complication in diabetic patients. Compared with conventional fundus color photography, fundus fluorescein angiography can dynamically display retinal vessel permeability changes, offering unique advantages in detecting early small lesions such as microaneurysms. However, existing intelligent diagnostic research on diabetic retinopathy images primarily focuses on fundus color photography, with relatively insufficient research on complex lesion recognition in fluorescein angiography images. This study proposed an adaptive multi-label classification model (D-LAM) to improve the recognition accuracy of small lesions by constructing a category-adaptive mapping module, a label-specific decoding module, and an innovative loss function. Experimental results on a self-built dataset demonstrated that the model achieved a mean average precision of 96.27%, a category F1-score of 91.21%, and an overall F1-score of 94.58%, with particularly outstanding performance in recognizing small lesions such as microaneurysms (AP = 1.00), significantly outperforming existing methods. The research provides reliable technical support for clinical diagnosis of diabetic retinopathy based on fluorescein angiography.
Accurate detection of cephalometric landmarks is crucial for orthodontic diagnosis and treatment planning. Current landmark detection methods are mainly divided into heatmap-based and regression-based approaches. However, these methods often rely on parallel computation of multiple models to improve accuracy, significantly increasing the complexity of training and deployment. This paper presented a novel regression method that can simultaneously detect all cephalometric landmarks in high-resolution X-ray images. By leveraging the encoder module of Transformer, a dual-encoder model was designed to achieve coarse-to-fine localization of cephalometric landmarks. The entire model consisted of three main components: a feature extraction module, a reference encoder module, and a fine-tuning encoder module, responsible for feature extraction and fusion of X-ray images, coarse localization of cephalometric landmarks, and fine localization of landmarks, respectively. The model was fully end-to-end differentiable and could learn the intercorrelation relationships between cephalometric landmarks. Experimental results showed that the successful detection rate (SDR) of our algorithm was superior to other existing methods. It attained the highest 2 mm SDR of 89.51% on test set 1 of the ISBI2015 dataset and 90.68% on the test set of the ISBI2023 dataset. Meanwhile, it reduces memory consumption and enhances the model's popularity and applicability, providing more reliable technical support for orthodontic diagnosis and treatment plan formulation.
Thyroid nodules are a common endocrine disorder, and their early detection and accurate diagnosis are crucial for the prevention of thyroid cancer. However, the highly heterogeneous morphology and boundaries of thyroid nodules pose significant challenges to their precise identification and classification. Traditional diagnostic approaches rely heavily on physicians' experience, which increases the risk of misdiagnosis and missed diagnoses. With the rapid advancement of computer-aided diagnosis (CAD) technologies, applying deep learning algorithms to the analysis of thyroid nodule ultrasound images has shown great potential. This paper reviews the latest research progress on deep learning-based CAD methods for thyroid nodules, with a focus on their applications in image preprocessing, segmentation and classification. The advantages and limitations of current techniques are analyzed, and potential future directions are discussed. This review aims to highlight the potential of deep learning in thyroid nodule diagnosis and to provide a foundation for selecting feasible pathways for future clinical applications.
