In bone tissue engineering microstructure design, adjusting the structural design of biomimetic bone scaffolds can provide distinct differentiation stimuli to cells on the scaffold surface. This study explored the biomechanical impacts of different biomimetic microstructures on advanced bone tissue engineering scaffolds. Two irregular bone scaffolds (homogeneous/radial gradient) based on the Voronoi tesselation algorithm and eight regular lattice scaffolds involving pillar body centered cubic, vintiles, diamond, and cube (homogeneous/radial gradient) with constant 80% porosity were constructed. Mechanical stimulation differentiation algorithms, finite element analysis, and computational fluid dynamics were used to investigate the effects of different pore structures on the octahedral shear strain and fluid flow shear stress within the scaffolds, thereby elucidating the differentiation capabilities of the five structural bone/cartilage cell types. The findings demonstrated that irregular structures and radial-gradient designs promoted osteogenic differentiation, whereas regular structures and homogeneous designs facilitated chondrogenic differentiation. The highest percentages of osteoblast and chondrocyte differentiation were observed in radial-gradient irregular scaffolds. This research provides insights into the microstructure design of bone tissue engineering implants.

Frequency-modulated continuous-wave radar enables the non-contact and privacy-preserving recognition of human behavior. However, the accuracy of behavior recognition is directly influenced by the spatial relationship between human posture and the radar. To address the issue of low accuracy in behavior recognition when the human body is not directly facing the radar, a method combining local outlier factor with Doppler information is proposed for the correction of multi-classifier recognition results. Initially, the information such as distance, velocity, and micro-Doppler spectrogram of the target is obtained using the fast Fourier transform and histogram of oriented gradients - support vector machine methods, followed by preliminary recognition. Subsequently, Platt scaling is employed to transform recognition results into confidence scores, and finally, the Doppler - local outlier factor method is utilized to calibrate the confidence scores, with the highest confidence classifier result considered as the recognition outcome. Experimental results demonstrate that this approach achieves an average recognition accuracy of 96.23% for comprehensive human behavior recognition in various orientations.

Magnetic tracking technologies have a promising application in detecting the real-time position and attitude of a capsule endoscope. However, most of them need to measure the magnetic moment of a permanent magnet (PM) embedded in the capsule accurately in advance, which can cause inconvenience to practical application. To solve this problem, this paper proposes a magnetic tracking system with the capability of measuring the magnetic moment of the PM automatically. The system is constructed based on a 4 × 4 magnetic sensor array, whose sensing data is analyzed to determine the magnetic moment by referring to a magnetic dipole model. With the determined magnetic moment, a method of fusing the linear calculation and Levenberg-Marquardt algorithms is proposed to determine the 3D position and 2D attitude of the PM. The experiments verified that the proposed system can achieve localization errors of 0.48mm, 0.42mm, and 0.83mm and orientation errors of 0.66 ◦ , 0.64 ◦ , and 0.87◦ for a PM (∅10mm × 10mm) at vertical heights of 5 cm, 10 cm, and 15 cm from the magnetic sensor array, respectively.

Master robots are integral components of teleoperated robot-assisted minimally invasive surgery systems. Among them, parallel mechanism-based 6-degree-of-freedom master robots are distinguished by low inertia and high-force feedback. However, complex kinematics and singularities are the main barriers limiting its usage. This study converts the Hexa-type 6-RUS mechanism into a master robot to construct master-slave teleoperation system. The clinical background is briefly introduced and a representative surgical robot is employed to analyze the master-slave mapping relationship. The inverse/forward kinematics, the Jacobian matrix, and the translation and orientation workspace are derived as the bases of master robot’s application. The architecture parameters are optimized by the global transmission index to achieve better motion/force transmissibility. Based on the optimal result, the prototype and the master-slave control loop are constructed. Finally, the corresponding master-slave teleoperation experiment and model experiment demonstrate that the proposed master robot satisfies the basic need for medical application.

Segmentation of the retinal vessels in the fundus is crucial for diagnosing ocular diseases. Retinal vessel images often suffer from category imbalance and large scale variations. This ultimately results in incomplete vessel segmentation and poor continuity. In this study, we propose CT-MFENet to address the aforementioned issues. First, the use of context transformer (CT) allows for the integration of contextual feature information, which helps establish the connection between pixels and solve the problem of incomplete vessel continuity. Second, multi-scale dense residual networks are used instead of traditional CNN to address the issue of inadequate local feature extraction when the model encounters vessels at multiple scales. In the decoding stage, we introduce a local-global fusion module. It enhances the localization of vascular information and reduces the semantic gap between high- and low-level features. To address the class imbalance in retinal images, we propose a hybrid loss function that enhances the segmentation ability of the model for topological structures. We conducted experiments on the publicly available DRIVE, CHASEDB1, STARE, and IOSTAR datasets. The experimental results show that our CT-MFENet performs better than most existing methods, including the baseline U-Net.

Addressing challenges such as low performance, high data signal-to-noise ratio requirements, and limited real-time capabilities in existing heart rate detection methods based on millimeter wave radar, this study presents a heart rate sensing approach tailored for weak vital sign signals characterized by low signal-to-noise ratio and missing data. The method applies a signal mask for echo sequences with variable length. Building upon this signal mask, a signal mapping technique that leverages morphology is devised to mitigate interference and noise. Additionally, learnable position encoding is incorporated to capture temporal features within the signal. Subsequently, a transformer encoder module is employed for matching and computation, culminating in the development of a time-series global regression model based on deep learning framework. Following the preparation of the dataset and model training, the proposed approach is validated by performance analysis experiments, interference resistance tests, and comparative experiments. Results indicate that this method achieves an impressive accuracy of 96.30% within signal durations ranging from 2 s to 5 s, and it is suitable for scenarios involving missing data and noise interference. Importantly, this approach effectively enables a precise heart rate sensing from short-duration radar signals.

Physiological signals such as electroencephalogram (EEG) signals are often corrupted by artifacts during the acquisition and processing. Some of these artifacts may deteriorate the essential properties of the signal that pertains to meaningful information. Most of these artifacts occur due to the involuntary movements or actions the human does during the acquisition process. So, it is recommended to eliminate these artifacts with signal processing approaches. This paper presents two mechanisms of classification and elimination of artifacts. In the first step, a customized deep network is employed to classify clean EEG signals and artifact-included signals. The classification is performed at the feature level, where common space pattern features are extracted with convolutional layers, and these features are later classified with a support vector machine classifier. In the second stage of the work, the artifact signals are decomposed with empirical mode decomposition, and they are then eliminated with the proposed adaptive thresholding mechanism where the threshold value changes for every intrinsic mode decomposition in the iterative mechanism.

The vagus nerve plays a pivotal role in regulating blood pressure, making vagus nerve stimulation a promising therapy for refractory hypertension. Nevertheless, most current research on vagus nerve stimulation for hypertension regulation employs rigid electrodes outside the nerve bundle, with limited exploration into the electrical stimulation paradigms. In this study, we employed the carbon nanotube yarn electrode, a flexible electrode, implanted in the left vagus nerve of rats to compare the modulatory effects of duty cycle and pulse width stimulation paradigms. Furthermore, we conducted a quantitative electrical stimulation experiment using the optimized duty cycle paradigm. The result showed that low-frequency stimulation yielded superior blood pressure regulation, whereas high-frequency stimulation resulted in apnea. In conclusion, intrafascicular vagus nerve stimulation with the duty-cycle paradigm demonstrated superior efficacy in reducing blood pressure compared to the pulse-width paradigm, with an optimal duty cycle identified at 20%. These findings offer valuable insights for optimizing vagus nerve stimulation protocols in the treatment of hypertension.

CircRNAs, widely found throughout the human bodies, play a crucial role in regulating various biological processes and are closely linked to complex human diseases. Investigating potential associations between circRNAs and diseases can enhance our understanding of diseases and provide new strategies and tools for early diagnosis, treatment, and disease prevention. However, existing models have limitations in accurately capturing similarities, handling the sparse and noise attributes of association networks, and fully leveraging bioinformatical aspects from multiple viewpoints. To address these issues, this study introduces a new non-negative matrix factorization-based framework called NMFMSN. First, we incorporate circRNA sequence data and disease semantic information to compute circRNA and disease similarity, respectively. Given the sparse known associations between circRNAs and diseases, we reconstruct the network to complete more associations by imputing missing links based on neighboring circRNA and disease interactions. Finally, we integrate these two similarity networks into a non-negative matrix factorization framework to identify potential circRNA-disease associations. Upon conducting 5-fold cross-validation and leave-one-out cross-validation, the AUC values for NMFMSN reach 0.971 2 and 0.976 8, respectively, outperforming the currently most advanced models. Case studies on lung cancer and hepatocellular carcinoma show that NMFMSN is a good way to predict new associations between circRNAs and diseases.

Automated sleep stages classification facilitates clinical experts in conducting treatment for sleep disorders, as it is more time-efficient concerning the analysis of whole-night polysomnography (PSG). However, most of the existing research only focused on public databases with channel systems incompatible with the current clinical measurements. To narrow the gap between theoretical models and real clinical practice, we propose a novel deep learning model, by combining the vision transformer with supervised contrastive learning, realizing the efficient sleep stages classification. Experimental results show that the model facilitates an easier classification of multi-channel PSG signals. The mean F1-scores of 79.2% and 76.5% on two public databases outperform the previous studies, showing the model’s great capability, and the performance of the proposed method on the children’s small database also presents a high mean accuracy of 88.6%. Our proposed model is validated not only on the public databases but the provided clinical database to strictly evaluate its clinical usage in practice.

The purpose of this study is to establish a multivariate nonlinear regression mathematical model to predict the displacement of tumor during brain tumor resection surgery. And the study will be integrated with augmented reality technology to achieve three-dimensional visualization, thereby enhancing the complete resection rate of tumor and the success rate of surgery. Based on the preoperative MRI data of the patients, a 3D virtual model is reconstructed and 3D printed. A brain biomimetic model is created using gel injection molding. By considering cerebrospinal fluid loss and tumor cyst fluid loss as independent variables, the highest point displacement in the vertical bone window direction is determined as the dependent variable after positioning the patient for surgery. An orthogonal experiment is conducted on the biomimetic model to establish a predictive model, and this model is incorporated into the augmented reality navigation system. To validate the predictive model, five participants wore HoloLens2 devices, overlaying the patient’s 3D virtual model onto the physical head model. Subsequently, the spatial coordinates of the tumor’s highest point after displacement were measured on both the physical and virtual models (actual coordinates and predicted coordinates, respectively). The difference between these coordinates represents the model’s prediction error. The results indicate that the measured and predicted errors for the displacement of the tumor’s highest point on the X and Y axes range from .0.678 7mm to 0.295 7mm and .0.431 4mm to 0.225 3mm, respectively. The relative errors for each experimental group are within 10%, demonstrating a good fit of the model. This method of establishing a regression model represents a preliminary attempt to predict brain tumor displacement in specific situations. It also provides a new approach for surgeons. By combining augmented reality visualization, it addresses the need for predicting tumor displacement and precisely locating brain anatomical structures in a simple and cost-effective manner.

Drosophila melanogaster has been a popular model organism in the study of sleep and circadian rhythm. The Drosophila activity monitoring (DAM) system is one of the many tools developed for investigating sleep behavior in fruit flies and has been acknowledged by researchers around the world for its simplicity and cost-effectiveness. Based on the simple activity data collected by the DAM system, a wide range of parameters can be generated for sleep and circadian studies. However, current programs that analyze DAM data cover a limited number of metrics and fail to provide individual data for the user to plot graphs and conduct analysis using other software. Therefore, we have developed SleepyFlyR, an R package that: (1) is simple and easy to use with a user-friendly user interface script; (2) provides a comprehensive analysis of sleep and activity parameters; (3) generates double-plotted graphs for sleep and activity patterns; (4) offers visualization of sleep and activity profiles across multiple days or within a single day; (5) calculates the changes of sleep and activity parameters between baseline and experiment; (6) stores both summary data and individual data in files with unique title.

Chinese Wine Culture influences people’s attitudes toward alcohol. The current study focuses on exploring the main features of a localized digital alcohol consumption intervention mini-program to address the difficulties of abstinence for native drinkers, as a promising way for long-term management of rehabilitation from alcohol use disorder. A mixed-method approach was used in this study. The self-report quantitative questionnaire recruited three groups of participants: 89 drinkers, 67 drinkers’ relatives, and 30 medical staff. The focus group qualitative interview inspected 36 participants’ perspectives on the core topics, including 21 drinkers, 4 drinkers’ relatives, and 11 medical staff. The results of combining the quantitative study and qualitative study indicated that the top difficulties of abstinence for native drinkers are the strong craving from the inside, the environmental influence, and the psychological health status, especially emotional states. Correspondingly, the most desired main features in an alcohol consumption digital intervention tool are the daily track of drinking conditions and craving level, periodic feedback reports that can share with others, and mood improvement training. Moreover, the top factors that influence participants’ intention to use/recommend the tool are whether the tool is effective, whether the user experience is good, and whether the tool can replenish the deficiency of the current alcohol treatment. Future work needs to balance what patients want and what others around them expect, so that potential users can benefit best from the digital intervention tool in the context of Chinese culture.

The knee joint is structurally complex and there are numerous factors that influence knee dynamics. Therefore, it is valuable to study the effect of stride length on knee contact during walking. Moreover, it is crucial to study the mechanical properties of the knee joint for the protection of the knee joint and the mechanism of knee diseases. In this study, a healthy volunteer was invited to investigate the kinematics of the lower limb under different stride lengths by conducting motion capture experiments. Then, a complete and detailed finite element model of the knee was established, and the effect of stride length on the knee contact was studied using the finite element method, where the boundary conditions and loads were set up in accordance with the actual working conditions based on the data obtained from the motion capture experiments. When the stride length was increased by 23.08% compared with the habitual stride length, the knee flexion angle at the beginning moment of the single-legged support phase could be increased by 108.12%, the maximum von Mises stress values on the femur cartilage and meniscus were increased from 5.888 to 16.023MPa and from 5.599 to 17.387 MPa, respectively, and the high-stress zone on the contact surface was also significantly shifted. When the stride length was reduced by 12.31% compared to the habitual stride length, the knee flexion angle at the moment of the end of the singlelegged support phase was reduced by 62.22%, and the maximum von Mises stress values on the femur cartilage and meniscus were reduced from 5.362MPa to 2.074MPa and from 5.255MPa to 1.986MPa, respectively. The results of this paper indicate that when exercising and preventing or treating stride knee diseases by walking, people should choose a suitable stride for exercise according to the health condition of the knee and avoid over-pursuing a large stride to improve the exercise effect, while a smaller stride is suitable for most people.

Portal vein stenosis is one of the common complications after liver transplantation in children. Accurate hemodynamic assessment is crucial for predicting the risk of complications after liver transplantation. In order to predict the location of portal vein thrombosis after liver transplantation surgery, single-outlet and three-outlet vascular models were reconstructed from computed tomography images by commercial software MIMICS. The velocity field was measured using a 9.4T magnetic resonance imaging scanner. Based on the experiment data of magnetic resonance velocimetry, computational fluid dynamics was verified, validated and then used to study the pressure and shear stresses on the wall of the two portal vein models. The simulation results can serve for the clinical prediction of early thrombosis after liver transplantation in portal vein.

Surgical site infections (SSIs) are the most common healthcare-related infections in patients with lung cancer. Constructing a lung cancer SSI risk prediction model requires the extraction of relevant risk factors from lung cancer case texts, which involves two types of text structuring tasks: attribute discrimination and attribute extraction. This article proposes a joint model, Multi-BGLC, around these two types of tasks, using bidirectional encoder representations from transformers (BERT) as the encoder and fine-tuning the decoder composed of graph convolutional neural network (GCNN) + long short-term memory (LSTM) + conditional random field (CRF) based on cancer case data. The GCNN is used for attribute discrimination, whereas the LSTM and CRF are used for attribute extraction. The experiment verified the effectiveness and accuracy of the model compared with other baseline models.

Aimed at the issues of high feature dimensionality, excessive data redundancy, and low recognition accuracy of using single classifiers on ground-glass lung nodule recognition, a recognition method was proposed based on CatBoost feature selection and Stacking ensemble learning. First, the method uses a feature selection algorithm to filter important features and remove features with less impact, achieving the effect of data dimensionality reduction. Second, random forests classifier, decision trees, K-nearest neighbor classifier, and light gradient boosting machine were used as base classifiers, and support vector machine was used as meta classifier to fuse and construct the ensemble learning model. This measure increases the accuracy of the classification model while maintaining the diversity of the base classifiers. The experimental results show that the recognition accuracy of the proposed method reaches 94.375%. Compared to the random forest algorithm with the best performance among single classifiers, the accuracy of the proposed method is increased by 1.875%. Compared to the recent deep learning methods (ResNet+GBM+Attention and MVCSNet) on ground-glass pulmonary nodule recognition, the proposed method’s performance is also better or comparative. Experiments show that the proposed model can effectively select features and make recognition on ground-glass pulmonary nodules.

Medical image segmentation is a crucial task in clinical applications. However, obtaining labeled data for medical images is often challenging. This has led to the appeal of semi-supervised learning (SSL), a technique adept at leveraging a modest amount of labeled data. Nonetheless, most prevailing SSL segmentation methods for medical images either rely on the single consistency training method or directly fine-tune SSL methods designed for natural images. In this paper, we propose an innovative semi-supervised method called multi-consistency training (MCT) for medical image segmentation. Our approach transcends the constraints of prior methodologies by considering consistency from a dual perspective: output consistency across different up-sampling methods and output consistency of the same data within the same network under various perturbations to the intermediate features. We design distinct semi-supervised loss regression methods for these two types of consistencies. To enhance the application of our MCT model, we also develop a dedicated decoder as the core of our neural network. Thorough experiments were conducted on the polyp dataset and the dental dataset, rigorously compared against other SSL methods. Experimental results demonstrate the superiority of our approach, achieving higher segmentation accuracy. Moreover, comprehensive ablation studies and insightful discussion substantiate the efficacy of our approach in navigating the intricacies of medical image segmentation.

Aimed at the problem of the low computational efficiency of the existing urethral-valve reliability analysis, an efficient reliability analysis method of the wireless energy-transmitting urethral-valve (WETUV) was proposed. The method is called FTA-AK-SS, based on the active learning Kriging (AK) model, subset simulation (SS) algorithm, and fault tree analysis (FTA). According to the principle of FTA, we established the fault tree model of the WETUV to determine its minimum cut set and bottom events. Then we defined the random variables affecting its reliability. The U learning function was used to selectively add the sample points of random variables to update the initial Kriging surrogate model. At the same time, combined with the SS algorithm, the reliability and sensitivity analyses of the WETUV were realized. The result shows that compared with the traditional Monte Carlo simulation and FTA-Kriging-SS methods, the proposed method significantly improves the calculation efficiency of the WETUV under the premise of ensuring calculation accuracy. The reliability of the WETUV is greatly affected by the rubber pad’s aging, the receiver coil’s corrosion, and the position deviation. This study can provide a new way to realize a high-efficiency reliability calculation and analysis for urethral valves.

A two-dimensional rectangular solenoid transmitting coil is proposed to address the problem that the three-dimensional receiving coil occupies excessive space inside the capsule robot. The transmitting coil consists of two pairs of rectangular solenoid coils distributed radially along the human body. By changing the direction of current flow, it can generate a two-dimensional magnetic field covering the whole central plane. Firstly, the working mechanism of the wireless power transfer system is introduced, and then the spatial electromagnetic field generated by the transmitting coil is analyzed through both mathematical calculations and finite element simulations. Finally, an experimental platform is built to determine the optimal resonant frequency of the system and validate its feasibility based on the power transfer efficiency and the receiving power. The experimental results demonstrate that when the receiving coil is located at the center of the coil pair, the receiving power is 1 416mW and the power transfer efficiency is 3.96%. Additionally, when the receiving coil operates in the central plane, it can receive sufficient energy regardless of the orientation.