Skeletal Motion Research Articles

Articulation skills of singing based on the biomechanical coordination of throat muscles

With the development of vocal training in music art, singing skills have been continuously explored and innovated. At present, the articulation skills of singing mostly focus on the surface operations of pronunciation skills such as tongue position, mouth shape, and lip movements, lacking the consideration of throat muscles and their coordination in clear pronunciation. In addition, everyone’s pronunciation habits are different, resulting in the low clarity of pronunciation of traditional techniques. This paper used the biomechanical model to analyze the coordination of throat muscles and proposed a systematic and scientific singing pronunciation and articulation training method. The study first built a biomechanical model of throat muscles based on Mooney-Rivlin and CAD, and introduced FEA to perform dynamic simulation on the model to simulate the mechanical behavior of muscle groups under different vocalization states. Then, it took the pharyngeal bones and muscle groups as the research objects, constructed a multi-rigid body dynamics model, and established the dynamic relationship between muscle drive and skeletal movement. Finally, the paper designed personalized singing pronunciation and clarity techniques based on CI, muscle tension distribution, etc. The experiment took 60 healthy singers as subjects, set up an experimental group and multiple control groups, and explored the effectiveness of singing articulation clarity techniques from the perspective of throat muscle biomechanics and visual coordination. The experimental results showed that the pronunciation clarity of the experimental group reached 7 points and the STOI reached 0.84, while the traditional oral resonance training group only scored 5 points and the STOI reached 0.72. The experimental results show that the singing pronunciation clarity technique based on the biomechanical coordination of throat muscles can significantly improve the singing pronunciation clarity and enhance the live listening effect of the singing.

Optimizing Athlete Training and Injury Mitigation Using Fuzzy Information-Based Skeletal Motion Analysis

Optimizing Athlete Training and Injury Mitigation Using Fuzzy Information-Based Skeletal Motion Analysis

Integrating gait and speech dynamics methodologies for enhanced stuttering detection across diverse datasets

<span lang="EN-US">Stuttering manifests as involuntary interruptions in the fluency of speech, often involving repetitions, prolongations, or blocks of sounds or syllables. These disruptions can significantly impact effective communication and psychosocial well-being. This research introduces a comprehensive system for speech impairment detection and gait analysis. Speech impairment, with a primary focus on stammer recognition, presents a multifaceted challenge in the field of speech processing. Stammers can manifest in various forms and detecting them accurately is a complex task. Our proposed methodology revolves around the development of StEnsembleNet, a neural network designed to learn spectral features at the frame level, enabling precise and efficient identification of speech impediments. Additionally, we extend our system's capabilities to the domain of gait analysis, leveraging a novel adaptive graph topology convolution network (AGT-ConvNet) for skeletal motion and visually enhanced topological learning to adapt to diverse visual environments and enhance the recognition of gait patterns. This research not only contributes to the field of speech therapy but also offers potential applications in healthcare and motion analysis.</span>

GelMA encapsulating BMSCs-exosomes combined with interference screw or suture anchor promotes tendon-bone healing in a rabbit model

The tendon-bone junction (TBJ), a critical transitional zone where tendons and bones connect, is particularly prone to injury due to the forces from muscle contractions and skeletal movements. Once tendon-bone injuries occur, the complex original tissue structure is difficult to restore, increasing the risk of re-tear. In this study, we initially established a rabbit model of tendon-bone injury and treated it using either interference screw or suture anchor. Biomechanical testing demonstrated the maximum tension and strength of TBJ with interference screw fixation were superior. However, histologic and immunohistochemical results showed more tissue regeneration and expression of cartilage markers at the site of injury with suture anchor fixation. Moreover, Gelatin Methacryloyl encapsulated with exosomes from mesenchymal stem cell (GelMA-exosomes) were prepared, showing a consistent and stable exosome release characteristic. The combined application of GelMA-exosomes with either interference screws or suture anchors further enhanced the healing of tendon-bone injuries, which may be achieved by promoting cellular proliferation as well as regulating the decreased expression of local pro-inflammatory factors IL-1β, IL-6 and TNF-α and increased expression of anti-inflammatory factors IL-10 and TGF-β. This provides a viable therapeutic strategy to enhance tendon-bone healing.

Effect of suspension training intervention on the balance ability of juvenile deaf-mute patients

Maintaining balance is essential for daily activities; deficits in hearing-impaired adolescents can severely impact their academics and daily life. Suspension training is employed to enhance the coordination of multiple muscle groups within the body and ultimately to improve balance. A total of 105 students with non-genetic hearing impairments, comprising 65 boys and 40 girls, participated in the study. Balance training using the TRS suspension belt included exercises such as prone elbow brace, supine back brace, kneeling elbow brace, lateral elbow brace, prone hip flexion, and supine hip lift. Strength training using equipment involved exercises like single-leg suspension squat, flying bird brace, supine suspension arm flexion, seated static brace, and standing butterfly pinch chest. The main outcome measures included static balance tests (lambda footwork, Wolfson postural stress test), dynamic balance tests (functional forward reach test, balance beam walk test), and vestibule function tests (rotation-walk test). The results of the lambda footwork and Wolfson postural stress tests revealed highly significant differences between 8 weeks and 16 weeks of intervention (p < 0.01). The Functional Forward Stretch test demonstrated a significant difference between 8 weeks of intervention and 2 weeks of intervention at 16 weeks (p < 0.05). The balance beam walking test indicated a significant improvement at 8 weeks (p < 0.05) and a highly significant improvement at 16 weeks (p < 0.01). In the spin-walk test, no statistically significant differences were observed between the pre-experimental test, the 8-week intervention, and the 16-week intervention (p > 0.05). Leveraging the expropriation system to mediate skeletal muscle movement and muscle contraction coordination can improve balance. Suspension training significantly improved balance during activities requiring static non-interference maintenance. However, suspension training did not demonstrate a significant effect on improving vestibule function. Overall, suspension training was effective in enhancing balance in all patients with non-hereditary deafness.

TriHuman: A Real-time and Controllable Tri-plane Representation for Detailed Human Geometry and Appearance Synthesis

Creating controllable, photorealistic, and geometrically detailed digital doubles of real humans solely from video data is a key challenge in Computer Graphics and Vision, especially when real-time performance is required. Recent methods attach a neural radiance field (NeRF) to an articulated structure, e.g., a body model or a skeleton, to map points into a pose canonical space while conditioning the NeRF on the skeletal pose. These approaches typically parameterize the neural field with a multi-layer perceptron (MLP) leading to a slow runtime. To address this drawback, we propose TriHuman a novel human-tailored, deformable, and efficient tri-plane representation, which achieves real-time performance, state-of-the-art pose-controllable geometry synthesis as well as photorealistic rendering quality. At the core, we non-rigidly warp global ray samples into our undeformed tri-plane texture space, which effectively addresses the problem of global points being mapped to the same tri-plane locations. We then show how such a tri-plane feature representation can be conditioned on the skeletal motion to account for dynamic appearance and geometry changes. Our results demonstrate a clear step toward higher quality in terms of geometry and appearance modeling of humans as well as runtime performance.

Automated Microfluidics-Assisted Hydrogel-Based Wet-Spinning for the Biofabrication of Biomimetic Engineered Myotendinous Junction.

The muscle-tendon junction (MTJ) plays a pivotal role in efficiently converting the muscular contraction into a controlled skeletal movement through the tendon. Given its complex biomechanical intricacy, the biofabrication of such tissue interface represents a significant challenge in the field of musculoskeletal tissue engineering. Herein, a novel method to produce MTJ-like hydrogel yarns using a microfluidics-assisted 3D rotary wet-spinning strategy is developed. Optimization of flow rates, rotational speed, and delivery time of bioinks enables the production of highly compartmentalized scaffolds that recapitulate the muscle, tendon, and the transient MTJ-like region. Additionally, such biofabrication parameters are validated in terms of cellular response by promoting an optimal uniaxial alignment for both muscle and tendon precursor cells. By sequentially wet-spinning C2C12 myoblasts and NIH 3T3 fibroblasts, a gradient-patterned cellular arrangement mirroring the intrinsic biological heterogeneity of the MTJ is successfully obtained. The immunofluorescence assessment further reveals the localized expression of tissue-specific markers, including myosin heavy chain and collagen type I/III, which demonstrate muscle and tenogenic tissue maturation, respectively. Remarkably, the muscle-tendon transition zone exhibits finger-like projection of the multinucleated myotubes in the tenogenic compartment, epitomizing the MTJ signature architecture.

Enhanced 2D Hand Pose Estimation for Gloved Medical Applications: A Preliminary Model.

(1) Background: As digital health technology evolves, the role of accurate medical-gloved hand tracking is becoming more important for the assessment and training of practitioners to reduce procedural errors in clinical settings. (2) Method: This study utilized computer vision for hand pose estimation to model skeletal hand movements during in situ aseptic drug compounding procedures. High-definition video cameras recorded hand movements while practitioners wore medical gloves of different colors. Hand poses were manually annotated, and machine learning models were developed and trained using the DeepLabCut interface via an 80/20 training/testing split. (3) Results: The developed model achieved an average root mean square error (RMSE) of 5.89 pixels across the training data set and 10.06 pixels across the test set. When excluding keypoints with a confidence value below 60%, the test set RMSE improved to 7.48 pixels, reflecting high accuracy in hand pose tracking. (4) Conclusions: The developed hand pose estimation model effectively tracks hand movements across both controlled and in situ drug compounding contexts, offering a first-of-its-kind medical glove hand tracking method. This model holds potential for enhancing clinical training and ensuring procedural safety, particularly in tasks requiring high precision such as drug compounding.

In-silico evaluation of orthodontic miniscrew-assisted rapid palatal expanders for patients with various stages of skeletal maturation

This study aimed to use a finite element method to assess the stress on the miniscrews, skeletal and dental changes resulting from maxillary expansion using either a tooth-bone-borne (TBB) or a bone-borne (BB) device on patients with various skeletal maturation. Two types of expanders were modeled using SolidWorks. The design of the 2 expanders was similar, with the exception that the BB appliance lacked support for teeth (rods and bands). Both were placed on the palatal bones with the help of 4 miniscrews as bony anchorage. Five skeletal maturation stages were examined using suture maturation classification by Angelieri. A lateral displacement of 0.2 mm was applied to simulate 1 turn of jackscrew per day. The dental and skeletal changes from the treatment were quantified. Stage A sutural maturation exhibited the greatest skeletal movement (0.13 mm/d) with a V-shaped displacement pattern. Stages B and C exhibited a more parallel expansion pattern. Stages D and E exhibited the least amount of skeletal movement with either device. Compared with the BB device, TBB demonstrated greater dental displacement. On average, the BB device exhibited a skeletal-to-dental expansion ratio of 70% from stage A to C compared with 49% with the TBB device. BB device showed more skeletal displacement than TBB, and the latter showed more dental side effects, particularly during stages B and C. The miniscrew-assisted rapid palatal expansion appliance was effective with sutural maturation stages A to C using the classification by Angelieri with a more parallel separation of the midpalatal sutures in stages B and C.

Preparation of Rutin-Whey Protein Pickering Emulsion and Its Effect on Zebrafish Skeletal Muscle Movement Ability.

Nutritional supplementation enriched with protein and antioxidants has been demonstrated to effectively strengthen skeletal muscle function and mitigate the risk of sarcopenia. Dietary protein has also been a common carrier to establish bioactive delivery system. Therefore, in this study, a Pickering emulsion delivery system for rutin was constructed with whey protein, and its structural characteristics, bioaccessibility, and molecular interactions were investigated. In the in vivo study, zebrafish (n = 10 in each group), which have a high genetic homology to humans, were treated with dexamethasone to induce sarcopenia symptoms and were administered with rutin, whey protein and the Pickering emulsion, respectively, for muscle movement ability evaluation, and zebrafish treated with or without dexamethasone was used as the model and the control groups, respectively. Results showed that the Pickering emulsion was homogeneous in particle size with a rutin encapsulation rate of 71.16 ± 0.15% and loading efficiency of 44.48 ± 0.11%. Rutin in the Pickering emulsion exhibited a significantly higher bioaccessibility than the free form. The interaction forces between rutin and the two components of whey proteins (α-LA and β-LG) were mainly van der Waals forces and hydrogen bonds. After treatment for 96 h, the zebrafish in Picking emulsion groups showed a significantly increased high-speed movement time and frequency, an increased level of ATP, prolonged peripheral motor nerve length, and normalized muscular histological structure compared with those of the model group (p < 0.05). The results of this study developed a new strategy for rutin utilization and provide scientific evidence for sarcopenia prevention with a food-derived resource.

Comprehensive treatment approach for hemifacial microsomia: Integrating orthognathic surgery with sequential customized implantation

This study aimed to evaluate the clinical outcomes of combining orthognathic surgery with staged patient-specific implants (PSIs) for comprehensive craniofacial asymmetry reconstruction in adult patients with hemifacial microsomia (HFM). Six adults with HFM (1 Type I and 5 Type IIa) underwent orthognathic surgery to correct skeletal malocclusion and chin deviation. Sequential PSIs were implanted to address craniofacial asymmetry. Digital lateral cephalograms and cranial computed tomography scans were obtained at four time points: pre-orthognathic surgery (T0), within three months after orthognathic surgery (T1), one year after orthognathic surgery and just before personalized implantation (T2), and after personalized implantation (T3). Evaluation parameters included skeletal and dentoalveolar measures, occlusal cant, chin deviation, skeletal stability, and facial contour symmetry. At T1, no significant differences were observed in skeletal movements compared with planned surgical movements (p>0.05). Similarly, at T2, skeletal movements did not significantly differ from those observed at T1 (p>0.05), indicating surgical precision and stability. Analysis of skeletal and dentoalveolar parameters, occlusal cant, and chin deviation revealed significant increases in SNB, FH-NPo, and ST N vert-Pog at T1 compared to T0 (p<0.05), along with notable improvements in chin deviation and occlusal cant (p<0.05). Comparison of T2 to T1 showed no significant changes in SNB, FH-NPo, ST N vert-Pog, chin deviation, or occlusal cant (p>0.05), indicating substantial postoperative stability. After personalized implantation (T3), further significant improvements were observed in skeletal symmetry. Combining orthognathic surgery with staged PSIs effectively reconstructs craniofacial asymmetry in adult patients with HFM, achieving significant improvements in skeletal alignment, occlusal cant, and chin deviation, with stable outcomes over time.

Enhanced human motion detection with hybrid RDA-WOA-based RNN and multiple hypothesis tracking for occlusion handling

Enhanced human motion detection with hybrid RDA-WOA-based RNN and multiple hypothesis tracking for occlusion handling

Can the Upper Vermilion and the Nasolabial Fold Be Changed With Orthognathic Surgery?

Retrusive profiles show an appearance of aging with an under-projected vermilion and pronounced nasolabial folds due to deficient bone support. A study was made of the association between orthognathic surgery and changes in the nasolabial and vermilion areas in patients with retrusive profiles. A retrospective cohort study evaluated patients subjected to bimaxillary surgery according to the Barcelona Line (BL) protocol during 2021 at Teknon Medical Center (Barcelona, Spain). Subjects with craniofacial syndromes, facial esthetic procedures, and dental rehabilitations involving lip changes, were excluded. The predictor variable was the timing of cephalometric measures, reported as T0 (preoperatively), T1 (1month after surgery), and T2 (after 1year of follow-up). The outcome variable corresponded to the soft tissue changes of the nasolabial and vermilion area, reported as the nasolabial fold length and angle, nasolabial angle, upper lip concavity, vermilion length, and upper lip sagittal distance from BL. The covariates comprised patient demographic data, the surgical-orthodontic protocol, and the magnitude and direction of the skeletal movements. Descriptive and inferential analyses were performed based on analysis of variance, the Bonferroni test, Pearson's linear coefficient, the nonparametric Mann-Whitney U-test, Kruskal-Wallis test, and multiple linear regression models. Statistical significance was considered for P<.05. The sample comprised 27 subjects with a mean age of 32.5±11.2years. A mean decrease in nasolabial angle of 5.5 ± 6° was recorded (P<.001), with a shortening of the nasolabial fold length of 4.4±7.6mm (P=.019). An increase in upper lip concavity angle of 14.4±12° was recorded (P<.001), along with a vermilion lengthening of 1.6±1.3mm (P<.001) and an increase in upper lip sagittal distance to BL of 5.7±7.3mm (P=.001), indicating a more projected and everted upper vermilion. When adequate dentoskeletal support is provided by specific positional changes of the jaws planned through orthognathic surgery, the length of the nasolabial fold decreases, and the upper vermilion lengthens and becomes slightly everted.

Attention-Based Variational Autoencoder Models for Human-Human Interaction Recognition via Generation.

The remarkable human ability to predict others' intent during physical interactions develops at a very early age and is crucial for development. Intent prediction, defined as the simultaneous recognition and generation of human-human interactions, has many applications such as in assistive robotics, human-robot interaction, video and robotic surveillance, and autonomous driving. However, models for solving the problem are scarce. This paper proposes two attention-based agent models to predict the intent of interacting 3D skeletons by sampling them via a sequence of glimpses. The novelty of these agent models is that they are inherently multimodal, consisting of perceptual and proprioceptive pathways. The action (attention) is driven by the agent's generation error, and not by reinforcement. At each sampling instant, the agent completes the partially observed skeletal motion and infers the interaction class. It learns where and what to sample by minimizing the generation and classification errors. Extensive evaluation of our models is carried out on benchmark datasets and in comparison to a state-of-the-art model for intent prediction, which reveals that classification and generation accuracies of one of the proposed models are comparable to those of the state of the art even though our model contains fewer trainable parameters. The insights gained from our model designs can inform the development of efficient agents, the future of artificial intelligence (AI).

Consensus for experimental design in electromyography (CEDE) project: Application of EMG to estimate muscle force

Skeletal muscles power movement. Deriving the forces produced by individual muscles has applications across various fields including biomechanics, robotics, and rehabilitation. Since direct in vivo measurement of muscle force in humans is invasive and challenging, its estimation through non-invasive methods such as electromyography (EMG) holds considerable appeal. This matrix, developed by the Consensus for Experimental Design in Electromyography (CEDE) project, summarizes recommendations on the use of EMG to estimate muscle force. The matrix encompasses the use of bipolar surface EMG, high density surface EMG, and intra-muscular EMG (1) to identify the onset of muscle force during isometric contractions, (2) to identify the offset of muscle force during isometric contractions, (3) to identify force fluctuations during isometric contractions, (4) to estimate force during dynamic contractions, and (5) in combination with musculoskeletal models to estimate force during dynamic contractions. For each application, recommendations on the appropriateness of using EMG to estimate force and justification for each recommendation are provided. The achieved consensus makes clear that there are limited scenarios in which EMG can be used to accurately estimate muscle forces. In most cases, it remains important to consider the activation as well as the muscle state and other biomechanical and physiological factors- such as in the context of a formal mechanical model. This matrix is intended to encourage interdisciplinary discussions regarding the integration of EMG with other experimental techniques and to promote advances in the application of EMG towards developing muscle models and musculoskeletal simulations that can accurately predict muscle forces in healthy and clinical populations.

Development and validation of a measure instrument for the understanding of structural-functional correlations of the locomotor system (MUSCLS)

ABSTRACT Learning new concepts in biology is closely related to previously learned concepts. For researchers, teachers and for educational authorities, differentiated knowledge of central conceptions in a field can help to develop, teach and evaluate new learning opportunities or educational approaches. Pre-conceptions have been investigated for numerous biological concepts in recent decades. One comparatively little studied area is that of skeletal movement. This article presents the development and validation of a test on the correlation between structure and function in the movement of extremities. The test was developed for pupils between the ages of 10 and 14, since skeletal movement is primarily a topic in schools at this age. The test comprises 10 items distributed over 2 scales. Validity is confirmed by confirmatory factor analysis, by high correlation coefficients with Science/Biology marks, and by further sources of validity evidence. Investigations of correlations between the test score achieved and personal parameters (gender and grade level) do not show statistically significant correlations, which indicates the fairness of the instrument. With a test completion time of 10 minutes, the knowledge test is time-efficient. The test extends the range of available assessment instruments on biological concepts in the area of the locomotor system.

Correlation of Orthognathic Surgical Movements to Perception of Facial Appearance in Patients With Cleft Lip and Palate.

Cephalometric scans were compared before and after surgery to assess the degree of correction. Correlations between skeletal movements and survey outcomes were determined using multivariate regression analysis. This study aims to identify relationships between subjective observer-reported improvements in esthetics and emotional appearance with specific surgical movements. Ten patients at a single tertiary institution (average age: 18.1 ± 0.8), 9 males and 1 female, underwent orthognathic repair and had comprehensive cephalometric records. Standardized anterior posterior and lateral pre and postoperative photographs of patients were included in a survey to clinicians to assess noncognitive domains on a Likert Scale (1-10). CLEFT-Q was administered to gauge patient satisfaction in categories of appearance, speech, and quality of life. Per clinicians, multiple domains increased including facial attractiveness (4.1 ± 0.7 versus 7.3 ± 0.7, P < 0.001), friendliness (4.5 ± 0.4 versus 7.3 ± 0.5, P < 0.001), confidence (4.1 ± 0.4 versus 7.1 ± 0.4, P < 0.001), and recommendation for surgery decreased (8.9 ± 0.1 versus 3.6 ± 0.5, P < 0.001). Speech distress decreased with increased SNA and convexity, whereas Psychological and Social scores decreased with an increased ANB. Functional eating and drinking scores increased with maxillary depth. Orthognathic surgery improves many noncognitive domains in patients with cleft lip and palate as assessed by both patients and clinicians on all aspects of facial attractiveness and perception. These findings demonstrate objective bases of skeletal adjustments for perceived improvements in facial appearance and emotion.

Визуализация и классификация движений человека на основе скелетной структуры: Нейросетевой подход к анализу спортивных упражнений и сравнение методологий

The authors of the paper review and compare different existing approaches to Human Action Recognition (HAR), analyze the advantages and disadvantages of platforms for extracting human skeletal structure from video stream, and evaluate the importance of visual representation in the motion analysis process. This paper presents an example implementation of one of the approaches to HAR based on the use of interpretability and visual expressiveness inherent in skeletal structures. In this work, an ad hoc network with Long Short-Term Memory (LSTM) for human activity classification is designed and implemented, which has been trained and tested in the domain of sports exercises. LSTM incorporation of memory cells and gating mechanisms not only mitigates the vanishing gradient problem but also enables LSTMs to selectively retain and utilize relevant information over extended sequences, making them highly effective in tasks with complex temporal dependencies. The problem with a fading gradient is quite common in deep neural networks and is that if the error is back propagated during the training of the network, the gradient can decrease strongly as it travels through the layers of the network to the initial layers. This can lead to the fact that the weights in the initial layers are practically not updated, which makes training of these layers impossible or slows down its process. The resulting solution can be used to create a real-time virtual fitness assistant. The resulting solution can be used to create a real-time virtual fitness assistant. In addition, this approach will make it possible to create interactive training applications with visualization of human skeletal structure, motion analysis and monitoring systems in the field of medicine and rehabilitation, as well as for the development of security systems with access control based on the analysis of visual data on the movement of human body parts.

Identification of different myofiber types in pigs muscles and construction of regulatory networks

BackgroundSkeletal muscle is composed of muscle fibers with different physiological characteristics, which plays an important role in regulating skeletal muscle metabolism, movement and body homeostasis. The type of skeletal muscle fiber directly affects meat quality. However, the transcriptome and gene interactions between different types of muscle fibers are not well understood.ResultsIn this paper, we selected 180-days-old Large White pigs and found that longissimus dorsi (LD) muscle was dominated by fast-fermenting myofibrils and soleus (SOL) muscle was dominated by slow-oxidizing myofibrils by frozen sections and related mRNA and protein assays. Here, we selected LD muscle and SOL muscle for transcriptomic sequencing, and identified 312 differentially expressed mRNA (DEmRs), 30 differentially expressed miRNA (DEmiRs), 183 differentially expressed lncRNA (DElRs), and 3417 differentially expressed circRNA (DEcRs). The ceRNA network included ssc-miR-378, ssc-miR-378b-3p, ssc-miR-24-3p, XR_308817, XR_308823, SMIM8, MAVS and FOS as multiple core nodes that play important roles in muscle development. Moreover, we found that different members of the miR-10 family expressed differently in oxidized and glycolytic muscle fibers, among which miR-10a-5p was highly expressed in glycolytic muscle fibers (LD) and could target MYBPH gene mRNA. Therefore, we speculate that miR-10a-5p may be involved in the transformation of muscle fiber types by targeting the MYHBP gene. In addition, PPI analysis of differentially expressed mRNA genes showed that ACTC1, ACTG2 and ACTN2 gene had the highest node degree, suggesting that this gene may play a key role in the regulatory network of muscle fiber type determination.ConclusionsWe can conclude that these genes play a key role in regulating muscle fiber type transformation. Our study provides transcriptomic profiles and ceRNA interaction networks for different muscle fiber types in pigs, providing reference for the transformation of pig muscle fiber types and the improvement of meat quality.

Transformative skeletal motion analysis: optimization of exercise training and injury prevention through graph neural networks.

Exercise is pivotal for maintaining physical health in contemporary society. However, improper postures and movements during exercise can result in sports injuries, underscoring the significance of skeletal motion analysis. This research aims to leverage advanced technologies such as Transformer, Graph Neural Networks (GNNs), and Generative Adversarial Networks (GANs) to optimize sports training and mitigate the risk of injuries. The study begins by employing a Transformer network to model skeletal motion sequences, facilitating the capture of global correlation information. Subsequently, a Graph Neural Network is utilized to delve into local motion features, enabling a deeper understanding of joint relationships. To enhance the model's robustness and adaptability, a Generative Adversarial Network is introduced, utilizing adversarial training to generate more realistic and diverse motion sequences. In the experimental phase, skeletal motion datasets from various cohorts, including professional athletes and fitness enthusiasts, are utilized for validation. Comparative analysis against traditional methods demonstrates significant enhancements in specificity, accuracy, recall, and F1-score. Notably, specificity increases by ~5%, accuracy reaches around 90%, recall improves to around 91%, and the F1-score exceeds 89%. The proposed skeletal motion analysis method, leveraging Transformer and Graph Neural Networks, proves successful in optimizing exercise training and preventing injuries. By effectively amalgamating global and local information and integrating Generative Adversarial Networks, the method excels in capturing motion features and enhancing precision and adaptability. Future research endeavors will focus on further advancing this methodology to provide more robust technological support for healthy exercise practices.