Cervical spondylotic myelopathy (CSM) and parkinsonian syndromes (PS) present similar motor symptoms, often causing misdiagnosis due to current clinical diagnostic limitations. Misdiagnosis can exacerbate patient conditions or result in unnecessary surgical interventions, thereby increasing surgical risks and the likelihood of serious postoperative complications. This study aims to develop a mixed dual-branch network for classifying CSM patients, PS patients, and healthy individuals using gait data. This study recruits 51 CSM patients, 49 PS patients, and 33 healthy controls. The kinematic data are collected and used to calculate the time series of angle, angular velocity, and angular acceleration for the hip, knee, and ankle joints. From each time series, 20 features are extracted, including the time domain, frequency domain, time-frequency domain, and nonlinear features. A dual-branch model named DCDM-Net is proposed to classify subjects through collaborative decision making (CDM) method, with one branch using ResNet with convolutional block attention module (CBAM) and evidential deep learning (EDL) loss for analyzing time series, and the other employing multilayer perceptron (MLP) for dealing with multi-domain features. DCDM-Net achieves an ACC of 92.35% $\pm ~0.76$ % and an AUC of 96.70% $\pm ~0.47$ % in the three-class classification task. Additionally, in binary classification scenarios, the model demonstrates robust performance with an average ACC of 93.13% and AUC of 98.34%. Furthermore, comparative evaluations show that the integrated EDL module surpasses Softmax, MC-Dropout, and Deep Ensembles in uncertainty estimation, yielding the lowest Expected Calibration Error (ECE of 0.0304) and lower Brier score (0.1074), indicating superior reliability. However, cross-dataset OOD validation yielded an AUROC of $0.4022~\pm ~0.2481$ and an AUPR of $0.9699~\pm ~0.0162$ , revealing that restricting features to joint angles leads to significant distribution overlap; this conversely validates that angular velocity and acceleration are indispensable for preventing model overconfidence. Interpretable results obtained through the SHapley Additive exPlanations (SHAP) method and the integrated gradients (IG) method are confirmed by clinical findings. Our method provides a promising tool for diagnosing CSM and PS, with the potential to reduce misdiagnosis. The code implementation of this study is available at https://github.com/AImedcinesdu212/DCDM-Net.

The disparity between image and text representations, often referred to as the modality gap, remains a significant obstacle for Vision Language Models (VLMs) in medical image segmentation. This gap complicates multi-modal fusion, thereby restricting segmentation performance. To address this challenge, we propose Evidence-driven Vision Language Model (EviVLM)—a novel paradigm that integrates Evidential Learning (EL) into VLMs to systematically measure and mitigate the modality gap for enhanced multi-modal fusion. To drive this paradigm, an Evidence Affinity Map Generator (EAMG) is proposed to collect complementary cross-modal evidences by learning a global cross-modal affinity map, thus refining modality-specific evidence embedding. An Evidence Differential Similarity Learning (EDSL) is further proposed to collect consistent cross-modal evidences by performing Bias-Variance Decomposition on differential matrix derived from bidirectional similarity matrices between image and text evidence embeddings. Finally, the subjective logic is used for mapping the collected evidences to opinions, and the Dempster-Shafer’s theory based combination rule is introduced for opinion aggregation, thereby quantifying the modality gap and facilitating effective multi-modal integration. Experimental results on three public medical image segmentation datasets validate that the proposed EviVLM can achieve state-of-the-art performance. Code is available at: https://github.com/QingtaoPan/EviVLM.

Multiple instance learning (MIL) is a crucial paradigm addressing weakly supervised classification in histopathological images. However, existing MIL methods struggle to model tile/patch interactions, which can capture important contextual information. Moreover, MIL is limited by suboptimal bag embeddings, as traditional methods focus primarily on extracting distinct embeddings for individual instances rather than for the entire bag. These limitations restrict MIL’s ability to leverage contextual information and generate discriminative aggregated representations. To address these limitations, we propose Interactive Contrastive Multiple Instance Learning (ICMIL), a novel MIL framework that integrates graph learning (GL) and contrastive learning (CL) to enhance MIL’s contextual awareness and discriminative power. ICMIL introduces two key innovations: (i) Transformer-based Graph Attention (TransGAT) models comprehensive patch interactions by constructing fully connected graphs and generating long-range edge attention for information aggregation. This approach addresses the problem of limited patch interactions in MIL and boosts MIL’s contextual awareness. (ii) Reinforced Contrastive MIL (ReCMIL) refines the bag embedding space by selecting the most advantageous contrastive bag pairs using a policy network. ReCMIL addresses the problem of suboptimal bag embeddings, enabling MIL to generate discriminative aggregated embeddings. Experimental results demonstrate the superiority of our proposed ICMIL method over state-of-the-art approaches on four publicly available datasets, which span three anatomical sites and encompass five classification tasks (including both binary and multiclass classification). Furthermore, we extend ICMIL with features from a pretrained foundation model and achieve the best performance. Specifically, ICMIL achieves validation accuracies of 92.39% on CRC-DX, 86.86% on CRC-KR, 100.00% on BRACS (binary), 66.67% on BRACS (multiclass), and 96.04% on TCGA-Lung. These findings highlight the strong potential of integrating ICMIL with pretrained foundation models for histopathology image analysis. The code will be available at: https://github.com/JingjiaoLou/ICMIL.

In multi-object text-to-image (T2I) diffusion, ensuring semantic consistency between textual prompts and generated visual content is crucial for image synthesis. However, such consistency constraint is often underemphasized in the denoising process of diffusion models. Although token supervised diffusion models can mitigate this issue by learning object-wise consistency between the image content and object segmentation maps, it tends to suffer from the problems of segmentation map bias and semantic overlap conflict, especially when involving multiple objects. In this paper, we propose ELDiff, a new evidential learning-supervised T2I diffusion model, which leverages the advantages of uncertainty metric and conflict detection to enhance the fault tolerance of unreliable segmentation maps and suppress semantic conflicts, strengthening object-wise consistency learning. Specifically, a pixel evidence loss is proposed to restrain overconfidence in unreliable labels through evidential regularization, and a token conflict loss is designed to weaken the contradiction between semantics through optimizing a measured conflict factor. Extensive experiments show that our ELDiff outperforms existing training based and train-free based T2I diffusion models on SD v1.4, SD v2.1, SDXL, SD v3.5, and Qwen-Image, without requiring additional inference-time manipulations. Notably, ELDiff can be seamlessly extended to the existing training pipeline of T2I diffusion models. Code can be found at https://github.com/QingtaoPan/ELDiff.

In the domain of robot skill learning, efficiently transferring source domain strategies is crucial for enhancing the generalization of acquired skills. However, existing transfer learning methods lack thorough analysis of the high-dimensional feature similarity between source and target domains, leading to limitations in transfer effectiveness. Therefore, this article proposes a robot assembly skill transfer strategy based on deep metric learning (STS-DML) to promote effective transfer of multi-source domain tasks. By constructing a metric model, the proposed method can efficiently evaluate the similarity between target and source domain tasks using only a small amount of target domain data. This provides a basis for adaptive adjustment of multi-source domain knowledge in strategy transfer, enabling effective strategy transfer across different assembly objects and environments. Extensive experiments conducted in both simulated and real environments demonstrate that the STS-DML algorithm achieves superior performance in the aspect of skill transfer success rate and efficiency.

Vision Language Model (VLM) has great potential to enhance the quality of pseudo labels in semi-supervised spine segmentation by leveraging textual class prompts to generate segmentation map, but no one has studied it yet. Although promising, it lacks explicit constraints to ensure consistency between spine class prompts and spine unit region, resulting in unsatisfactory performance in multi-class segmentation map generation. In this paper, we propose CPS4, the first text-guided semi-supervised spine segmentation network using class prompts to enhance the quality of spine pseudo labels. Specifically, CPS4 is implemented through two training stages. (i) Class-specific consistency constrained VLM pretraining stage: we propose token- and pixel-level attention loss to optimize the consistency between class prompts and spine units, forcing the textual class prompt to be closely coupled with the target spine unit in the semantic space. (ii) Class Prompt driven semi-supervised spine segmentation stage: using the pretrained vision-text encoder, we derive each class-specific binary segmentation map for the unlabeled spine image and integrate them into an unified multi-class segmentation map, improving the quality of the spine pseudo label generated by the semi-supervised spine segmentation network. Experimental results show that our CPS4 achieves superior spine segmentation performance with Dice of 80.44%, only using 5% labeled data on the public spine segmentation dataset, surpassing popular semi-supervised learning and VLM methods. Our code will be available.

Multiple Myeloma (MM) is a hematologic malignancy characterized by abnormal plasma cell proliferation and bone destruction. Increasing evidence underscores the immune system's critical role in MM initiation, therapy and recurrence. Mathematical modeling offers a quantitative framework for analyzing such complex interactions. Existing MM models have provided insights into tumor dynamics but largely overlooked immune regulation. To address this gap, we develop a novel model explicitly incorporating Natural Killer (NK) cells, Cytotoxic T Lymphocytes (CTLs), Regulatory T cells (Tregs) and Cytokines Interleukin-2 (IL-2) and Interleukin-10 (IL-10). The model first integrates two immune evasion mechanisms, antigen escape and an immunosuppressive microenvironment, alongside competition among MM subtypes for limited resources. It also simulates Chimeric Antigen Receptor-T cell (CAR-T) therapy to examine how immunosuppression influences outcomes. Simulations reproduce immune-tumor interactions in the bone microenvironment, reveal that antigen loss exerts a stronger effect on immune escape than immunosuppression alone, and show that increased immune pressure accelerates evolution toward resistant subtypes. Enhancing IL-2 and IL-10 critically modulates CAR-T cell abundance and efficacy. Model predictions agree with experimental data, improving invasion kinetics accuracy over models without immune components. This framework offers a robust basis for studying MM-immune dynamics and advancing immunotherapy modeling.

Gait analysis offers a promising approach for disease diagnosis, as various neurological and musculoskeletal disorders often lead to distinct gait patterns that can be systematically analyzed to assist in accurate clinical decision-making. While deep learning methods have been widely applied to multivariate gait time series, challenges remain-particularly in extracting complex coupling relationships across and within joints, as well as mitigating interference caused by asynchronous feature extraction during multi-feature fusion. In this paper, the Global-Local Interaction Rebalancing Gait Network (GLIR-GaitNet) is proposed to solve these challenges, incorporating two novel modules. The Global-Local Joint Coupling Feature Extractor (GL-JCFE) module integrates a residual-based 2D local representation with a dynamic graph-based global modeling approach to capture comprehensive inter-and intra-joint coupling relationships. Within this module, the Multi-feature Cross Enhancement (MCE) is further introduced to strengthen feature complementarity across domains. Additionally, the Prototypical Interaction Rebalance (PIR) module enhances the consistency of feature distribution by introducing cross-domain similarity loss monitoring while reducing interference between feature extraction processes. Extensive experiments conducted on three self-collected datasets (CSM-PS, LDH-LSS and CSM-HC) and one publicly available dataset (HOA) demonstrate that GLIR-GaitNet significantly outperforms state-of-theart classification methods in terms of diagnostic average ACC and AUC. These results highlight the strong capability of the GLIR-GaitNet in distinguishing easily confused gait-related diseases and fine-grained severity grading. Our Code is available at this repository: https://github.com/ginasmithe/GLIR-GaitNet.

Background and Objectives: Multiple myeloma (MM) is a prevalent malignant hematologic tumor with no current cure, highlighting the pressing need for novel drug development. Biochemical approaches to drug development involve extensive and costly biological experiments, underscoring the urgency to integrate new technologies for enhanced efficiency. Materials and Methods: This article presents a mathematical model to assess MM drug efficacy, aiming to guide drug development strategies and improve drug development efficiency. The model employs ordinary differential equations to simulate interactions among osteoblasts, osteoclasts, and MM cells within the bone microenvironment, as well as changes in the NF-[Formula: see text]B and c-Jun N-terminal kinase (JNK) signaling pathways post-MM invasion. Our study investigates the role of the first-line drugs denosumab and bisphosphonates, which are clinical treatments for bone-related complications caused by MM, and compares the effects of the drugs DTP3 and Japonicone A (JA), currently under development for the treatment of MM. Results: Our findings regarding denosumab align with existing experimental data, thus validating the model. Furthermore, our study demonstrates that denosumab exhibits the same anti-MM effect as bisphosphonates but reveals that denosumab acts more quickly due to its direct action on RANKL through monoclonal antibodies, whereas bisphosphonates act more slowly by targeting osteoclasts. Additionally, both DTP3 and JA effectively inhibit MM cell proliferation, positioning them as potential therapeutic agents, with DTP3 displaying superior efficacy compared to JA. Conclusions: We utilize mathematical models as alternatives to biological models to investigate the effects of drugs. These models not only elucidate the mechanisms of action of clinically used medications but also enhance our understanding of drug utilization, providing valuable guidance for physicians in their decision-making. Additionally, for drugs in development, these models can assess both efficacy and side effects, informing the drug development process. They also facilitate comparisons between different drugs to identify less effective options, thereby reducing costs. For example, in the case of DTP3 and JA, where DTP3 proves superior, the models help reduce research and development (R&D) costs associated with JA, saving both resources and time.

BACKGROUND:Osteoarthritis (OA) of the hip is a progressive musculoskeletal disorder characterized by stiffness and limited passive range of motion. Hip OA patients experience mobility impairment and altered gait patterns when compared to healthy controls (HCs). Although various interventions have been designed to alleviate these symptoms, it is unclear if there is a reliable method to track biomechanical changes in patients with unilateral hip OA in a clinical setting. PURPOSE:The purpose of this study is to evaluate the efficacy of lower extremity kinematic gait data for detecting and rating the severity of unilateral hip OA using machine learning algorithms. METHODS:First, a feature extraction framework is developed to derive several discriminative spatiotemporal and nonlinear features from lower extremity kinematic gait data. These features reflect the subtle disparity in gait characteristics, and can serve as indicators to distinguish between groups. Afterwards, the Shapley Additive exPlanations (SHAP) method is applied for feature selection and dimensionality reduction, providing detailed explanations of each feature's contribution to classification performance. Second, a support vector machine (SVM) is used to classify gait patterns between unilateral hip OA patients and HCs. Finally, the effectiveness of this strategy is comprehensively validated on a publicly available gait dataset, containing 80 asymptomatic participants and 99 patients with unilateral hip OA, who are classified according to Grades 2, 3, and 4 of Kellgren and Lawrence (KL). RESULTS:Using a cross-validation scheme of 10-fold, the classification accuracy achieves 98.21% for hip OA detection (HCs vs hip OA patients) and 89.65% (HCs vs Grade2/3 vs Grade 4) and 87.54% (HCs vs Grade2 vs Grade 3 vs Grade 4) for severity rating. CONCLUSION:The results demonstrate superior performance compared to other up-to-date methods, suggesting that the proposed method can serve as a supplementary tool to the KL grading scale for hip OA detection and severity assessment in clinical practice. Gait analysis provides objective data on the patient's walking pattern and can detect subtle changes in gait that may not be apparent on a radiographic image. TRIAL REGISTRATION:ClinicalTrials. gov (NCT01907503). The registration date of the clinical trial is 17th July, 2013.

Multiple instance learning (MIL) has proven effective in classifying whole slide images (WSIs), owing to its weakly supervised learning framework. However, existing MIL methods still face challenges, particularly over-fitting due to small sample sizes or limited WSIs (bags). Pseudo-bags enhance MIL's classification performance by increasing the number of training bags. However, these methods struggle with noisy labels, as positive patches often occupy small portions of tissue, and pseudo-bags are typically generated by random splitting. Additionally, they face difficulties with non-discriminative instance embeddings due to the lack of domain-specific feature extractors. To address these limitations, we propose Phenotype Clustering Reinforced Multiple Instance Learning (PCR-MIL), a novel MIL framework that integrates clusteringbased pseudo-bags to improve MIL's noise robustness and the discriminative power of instance embeddings. PCR-MIL introduces two key innovations: (i) Phenotype Clustering-based Feature Selection (PCFS) selects relevant instance embeddings for prediction. It clusters instances into phenotype-specific groups, assigns positive instances to each pseudo-bag, and then uses Grad-CAM to select the most relevant positive embeddings. This approach mitigates noisy label challenges and enhances MIL's robustness to noise; (ii) Reinforced Feature Extractor (RFE) uses reinforcement learning to train an extractor based on selected clean pseudobags instead of noisy samples. This approach improves the discriminative power of extracted instance embeddings and enhances the feature representation capabilities of MIL. Experimental results on the publicly available BRACS and CRC-DX datasets demonstrate that PCR-MIL outperforms state-of-the-art methods. The code is available at: https:// github.com/JingjiaoLou/PCR-MIL.

Semi-supervised medical image segmentation (SSMIS) uses consistency learning to regularize model training, which alleviates the burden of pixel-wise manual annotations. However, it often suffers from error supervision from low-quality pseudo labels. Vision-Language Model (VLM) has great potential to enhance pseudo labels by introducing text prompt guided multimodal supervision information. It nevertheless faces the cross-modal problem: the obtained messages tend to correspond to multiple targets. To address aforementioned problems, we propose a Dual Semantic Similarity-Supervised VLM (DuSSS) for SSMIS. Specifically, 1) a Dual Contrastive Learning (DCL) is designed to improve cross-modal semantic consistency by capturing intrinsic representations within each modality and semantic correlations across modalities. 2) To encourage the learning of multiple semantic correspondences, a Semantic Similarity-Supervision strategy (SSS) is proposed and injected into each contrastive learning process in DCL, supervising semantic similarity via the distribution-based uncertainty levels. Furthermore, a novel VLM-based SSMIS network is designed to compensate for the quality deficiencies of pseudo-labels. It utilizes the pretrained VLM to generate text prompt guided supervision information, refining the pseudo label for better consistency regularization. Experimental results demonstrate that our DuSSS achieves outstanding performance with Dice of 82.52%, 74.61% and 78.03% on three public datasets (QaTa-COV19, BM-Seg and MoNuSeg).

Low-quality pseudo labels pose a significant obstacle in semi-supervised medical image segmentation (SSMIS), impeding consistency learning on unlabeled data. Leveraging vision-language model (VLM) holds promise in ameliorating pseudo label quality by employing textual prompts to delineate segmentation regions, but it faces the challenge of cross-modal alignment uncertainty due to multiple correspondences (multiple images/texts tend to correspond to one text/image). Existing VLMs address this challenge by modeling semantics as distributions but such distributions lead to semantic degradation. To address these problems, we propose Alignment-Multiplicity Aware Vision-Language Model (AMVLM), a new VLM pre-training paradigm with two novel similarity metric strategies. (i) Cross-modal Similarity Supervision (CSS) proposes a probability distribution transformer to supervise similarity scores across fine-granularity semantics through measuring cross-modal distribution disparities, thus learning cross-modal multiple alignments. (ii) Intra-modal Contrastive Learning (ICL) takes into account the similarity metric of coarse-fine granularity information within each modality to encourage cross-modal semantic consistency. Furthermore, using the pretrained AMVLM, we propose a pioneering text-guided SSMIS network to compensate for the quality deficiencies of pseudo-labels. This network incorporates a text mask generator to produce multimodal supervision information, enhancing pseudo label quality and the model’s consistency learning. Extensive experimentation validates the efficacy of our AMVLM-driven SSMIS, showcasing superior performance across four publicly available datasets. The code will be available at: https://github.com/QingtaoPan/AMVLM.

BACKGROUND AND OBJECTIVE:Osteoporosis is characterized by reduced bone mass and deterioration of bone structure, yet screening rates prior to fractures remain low. Given its high prevalence and severe consequences, developing an effective osteoporosis screening model is highly significant. However, constructing these screening models presents two main challenges. First, selecting representative slices from CT image sequences is challenging, making it crucial to filter the most indicative slices. Second, samples lacking complete modal data cannot be directly used in multimodal fusion, resulting in underutilization of available data and limiting the performance of the multimodal osteoporosis screening model. METHODS:In this paper, we propose a reinforcement learning-driven knowledge distillation-assisted multimodal model for osteoporosis screening. The model integrates demographic characteristics, routine laboratory indicators, and CT images. Specifically, our framework includes two novel components: 1) a deep reinforcement learning-based image selection module (DRLIS) designed to select representative image slices from CT sequences; and 2) a knowledge distillation-assisted multimodal model (KDAMM) that transfers information from single-modal teacher networks to the multimodal model, effectively utilizing samples with incomplete modalities. The codes are published on: https://github.com/AImedcinesdu212/Osteoporosis-Predictionhttps://github.com/Hidden-neurosis/osreoporosis.git. RESULTS:The proposed multimodal osteoporosis screening model achieves an accuracy of 88.65 % and an AUC of 0.9542, surpassing existing models by 2.85 % in accuracy and 0.0212 in AUC. Additionally, we demonstrate the effectiveness of each novelty within our framework. The SHAP values are calculated to assess the importance of demographic characteristics and routine laboratory test data. CONCLUSION:This paper presents a knowledge distillation-assisted multimodal model for opportunistic osteoporosis screening. The model incorporates demographic characteristics, routine laboratory indicators (including blood tests and urinalysis), and CT images. Extensive experiments, conducted on self-collected datasets, validate that the proposed framework achieves state-of-the-art performance.

A recent deterministic learning theory has achieved locally-accurate identification of unknown system dynamics. This article presents a novel application of deterministic learning theory to unsupervised learning for the first time. Specifically, a new time series clustering strategy with a dynamics-based similarity measure is proposed. Firstly, the dynamics knowledge learned from the time series is represented and stored in the form of constant weights through deterministic learning theory. Secondly, dynamical estimators constructed with the learned dynamics knowledge are used to generate recognition errors, forming a similarity measure matrix to characterize the dynamics-based similarity between time series. Finally, the clustering of time series data with different dynamical behaviors is achieved based on the K-medoids prototype according to the dynamics-based similarity measure matrix. To verify the effectiveness of the proposed method, a dynamical pattern dataset based on benchmark dynamical systems (e.g., Lorenz, Chen, and Lü systems) is also constructed. The experimental results on a synthetic dataset and two real datasets demonstrate that the proposed method is superior to other well-known clustering algorithms in the clustering task for dynamical systems.

In the previous aerodynamic analysis using a potential flow-based unsteady aerodynamic model for the flapping wing rotor (FWR), the average rotary velocity and flapping frequency together with the flapping and twist angles are usually taken as the FWR kinematics of motion. In the present study, an unsteady vortex ring method (UVRM) is developed to take the rotary velocity variation of the FWR during the flapping motion into account based on 3D unsteady potential flow theory. The UVRM is validated by comparing with the published results of a flapping wing and an FWR model. An FWR test model and the corresponding experimental platforms including a wing motion tracking system and a force measurement system are built to measure the FWR motion and associated forces. Three FWR test cases of different input voltages and kinematics of flapping motion are considered to measure the aerodynamic forces and compare with the UVRM results. The results show that the differences between the measured and pre-set flapping angles (-50 degrees- 20 degrees) for the FWR are negligible for all the cases, but the differences in the variation amplitudes of the measured twist angle and the corresponding pre-set rigid twist angles can be as much as about 15 degrees. Also, the FWR rotary velocity varies dramatically during a flapping cycle. For one of the cases with the -10 degrees-30 degrees pre-set twist angle and 4 V input voltage, the difference between the maximum and minimum rotary speed is about 1451 degrees/s, which makes a significant effect on the FWR aerodynamic performance. The results also show that the calculated lift forces using the UVRM and the measured real-time rotary speeds during a flapping motion are very close to the experimental results. While the calculated results by taking the FWR's average rotary speeds show apparent differences (up to about 25%) from the measured lift forces in most of the cases. Overall, the UVRM provides an efficient method for the FWR aerodynamic analysis with higher accuracy by taking the effect of variable rotary motion and twist angle of the FWR during a flapping cycle into account.

Automatic movement analysis utilizing surveillance video is believed to be an important and convenient way for timely delirium detection in an Intensive Care Unit (ICU). However, video-based delirium movement detection (DMD) faces inherent challenges: 1) Irregular movements with large differences in the four limbs of a patient; 2) similar movements in delirium and normal situations, and large movement variations between patients experiencing delirium. To address the challenges 1, this paper proposes a Long-Short-View Aware Multi-Agent Reinforcement Learning (LS-MARL) method to identify the most representative movement snippet for DMD, considering that the global state provided by the long-view is important for guiding the agent's decision-making, but ignored in existing MARL methods. The proposed LS-MARL has two novel designs. First, a novel Teacher Auxiliary Policy (TAP) is developed for direction preperception of the representative movement snippet. Second, a new reward mechanism, Team Intrinsic Reward (TIR), is introduced to quantify the contribution of each agent. Experiments demonstrate that the proposed LS-MARL method outperforms state-of-the-art methods. Furthermore, to handle the second challenge, a new Self-Adjusting Ensemble Learning (SAEL) strategy is built to adaptively integrate an optimal classifier combination from multidomain features, which further improves the performance of classification tasks in the proposed LS-MARL method.

In this study, the aerodynamic performance of flapping wings using a parallel motion was investigated and compared with the insect-like "fan-sweep" motion, and the effect of adding a slit to the wings was analyzed. First, numerical simulations were performed to analyze the wing aerodynamics of two flapping motions with equivalent stroke amplitudes over a range of pitching angles based on computational fluid dynamics (CFD). The simulation results indicated that flapping wings with a rapid and short parallel motion achieved better lift and efficiency than those of the fan-sweep motion while maintaining the same aerodynamic characteristics regarding stall delay and leading-edge vortices. For a parallel motion with a pitching angle of 25° and 100 mm stroke amplitude, the wings generated an average lift of 8.4 gf with a lift-to-drag ratio of 1.06, respectively, which were 1.8% and 26% greater than those of the fan-sweep motion with a corresponding 96° stroke amplitude. This situation was reversed when the pitching angle and stroke amplitude were increased to 45° and 144° for the fan-sweep motion, which was equivalent to the parallel motion with a 150 mm stroke amplitude. The slit effect in the parallel motion was also evaluated, and the CFD results indicated that a slit width of 1 mm (1/50 wing chord) increased the lift of the wing by approximately 27% in the case of the 150 mm stroke amplitude. Further, the slit width slightly influenced the lift and aerodynamic efficiency.

Objective. In recent years, artificial intelligence-based electrocardiogram (ECG) methods have been massively applied to myocardial infarction (MI). However, the joint analysis of static and dynamic features to achieve accurate and interpretable MI detection has not been comprehensively addressed. Approach. This paper proposes a simplified ensemble tree method with a joint analysis of static and dynamic features to solve this issue for MI detection. Initially, the dynamic features are extracted by modeling the intrinsic dynamics of ECG via dynamic learning in addition to extracting classical static features. Secondly, a two-stage feature selection strategy is designed to identify a few significant features, which substitute the original variables that are employed in constructing the ensemble tree. This approach enhances the discriminative ability by selecting significant static and dynamic features. Subsequently, this paper presents an interpretable classification method named StackTree by introducing a stacked ensemble scheme to modify the ensemble tree simplification algorithm. The representative rules of the raw ensemble trees are selected as the intermediate training data that is used to retrain a decision tree with performance close to that of the source ensemble model. Using this scheme, the significant precision and interpretability of MI detection are thus comprehensively addressed. Main results. The effectiveness of our method in detecting MI is evaluated using the Physikalisch-Technische Bundesanstalt (PTB) and clinical database. The findings suggest that our algorithm outperforms the traditional methods based on a single type of feature. Additionally, it is comparable to the conventional random forest, achieving 97.1% accuracy under the inter-patient framework on the PTB database. Furthermore, feature subsets trained on PTB are validated using the clinical database, resulting in an accuracy of 84.5%. The chosen important features demonstrate that both static and dynamic information have crucial roles in MI detection. Crucially, the proposed method provides clear internal workings in an easy-to-understand visual manner.

Cervical spondylotic myelopathy (CSM) is the main cause of cervical spinal cord dysfunction in adults, especially in middle-aged and elderly patients, which easily leads to gait disturbance. In the present study, we propose a dynamic method for the detection of CSM based on nonlinear dynamics of gait system and deterministic learning theory. First, a 3-dimensional (3D) gait analysis system is used to capture the walking locomotion from healthy controls (HCs) and patients with CSM. Discriminant kinematic gait features, including angles of hip and knee joints in the sagittal and coronal planes, are extracted based on statistical analysis and clinicians’ empirical investigation. Second, deterministic learning theory is used to model and identify nonlinear gait system dynamics of HCs and patients with CSM, which are approximated and stored in constant Radial Basis Function (RBF) neural networks (NN). The disparity of gait system dynamics between the two groups of participants is used for classification and detection of the presence of CSM by constructing a bank of dynamic estimators with constant RBF NN. Finally, experiments are carried out on the self-constructed CSM gait database to evaluate the performance of the proposed method, in which gait data from 45 CSM patients and 45 age-matched HCs are involved. By using 2-fold and leave-one-out cross-validation styles, the achieved average classification accuracy is reported to be 94.44 % and 95.56 % , respectively. The results demonstrate excellent performance and the proposed method has the potential to serve as a candidate for the automatic detection of CSM in clinical examination.