Search

Abstract Direct current optimal power flow (DC‐OPF) problems need to be solved more frequently to maintain safety and economic power system operation. Traditional solvers take too much time to get optimal results. To overcome it, a new self‐supervised augmented Lagrangian neural network (ALNN) is proposed to solve DC‐OPF problem. The proposed ALNN consists of two neural networks: the control net and the penalty net. The control net predicts active power of generators; the penalty net updates the Lagrangian multipliers. The equality constraints are embedded into the control net to guarantee no equality violations. The generalized reduced gradient method is used to reduce theviolations of inequality constraint. The effectiveness of the proposed model is demonstrated on IEEE 118‐bus. The results show that with the help of equality embedding, the equality constraints are always satisfied, which in turn improves the feasibility of ALNN. Compared to the state‐of‐art models, the proposed model has higher feasibility and less constraint violations without comprising optimality. What is more, most of the inactive constraints can be found during the training process and then they are used to speed up the post‐processing part.

Abstract The worldwide occurrence of wind droughts challenges the balance of power systems between energy production and consumption. Expanding inter‐day energy storage serves as a strategic solution, yet optimizing its capacity depends on accurately modeling future renewable energy uncertainties to avoid over‐ or under‐investment. Existing approaches that use the historical extreme scenario set (HESS) to represent future conditions are contentious due to potential inadequacies in forecasting future extreme scenarios (ESs), including those on a decadal or centennial scale. This study addresses the issue by proposing an advanced energy storage expansion framework that leverages Extreme Value Theory (EVT) and a novel Deep Generative Model, namely the Diffusion Model. To model the extremes in a principled way, this work leverages EVT to establish a severity‐probability mapping for wind droughts, guiding the training process of the Diffusion Model. This model excels in generating ESs that accurately reflect the distribution of real‐world extremes, thereby significantly enhancing the predictive capacity of HESS. Case studies on a real‐world power system confirm the method's capacity to generate high‐quality ESs, encompassing the most severe historical wind droughts not included in the training dataset, thereby facilitating resilient energy storage expansion against unforeseen extremes.

Abstract Aims/Introduction Diabetic cardiomyopathy (DCM) is a prevalent condition among individuals with diabetes, and is associated with a high mortality rate. The anti‐oxidant properties of Jing Huang or Polygonatum sibiricum polysaccharide (PSP) have been extensively used to treat diabetes‐related disorders; however, its potential effectiveness against DCM remains unknown. This study aimed to investigate PSP's therapeutic effects on DCM in an experimental diabetic mouse model. Materials and Methods To induce insulin resistance, mice were fed a high‐fat diet for 3 months, followed by intraperitoneal streptozotocin injection to induce slight hyperglycemia and develop DCM. Both DCM and control mice were given PSP orally for 3 weeks. Western blotting was used to detect the protein expressions of protein kinase G, C/EBP homologous protein, glucose‐regulated protein 78, phosphodiesterase type 5, protein kinase R‐like endoplasmic reticulum (ER) kinase, and phospho‐protein kinase R‐like endoplasmic reticulum kinase in heart tissue. Results The results showed a reduction in bodyweight and blood glucose levels in the PSP therapy group compared with DCM group. PSP also improved cardiac function and had a negligible effect on malondialdehyde activity. Furthermore, the findings showed that PSP alleviated ER and oxidative stress observed in DCM mice hearts, leading to the inhibition of cyclic guanosine monophosphate‐specific phosphodiesterase type 5 and cardiac cyclic guanosine monophosphate reactivation. Phosphodiesterase type 5 inhibition reduced high‐fat diet‐induced cardiac dysfunction and decreased ER stress. Conclusions PSP could effectively protect diabetic myocardium by inhibiting endoplasmic reticulum stress. These findings provide crucial insights into the potential of PSP to ameliorate DCM conditions in diabetic mice by decreasing ER and oxidative stress, and enhancing cyclic guanosine monophosphate protein kinase G signaling.

Abstract The complement system, comprising over 30 proteins, is integral to the immune system, and the coagulation system is critical for vascular homeostasis. The activation of the complement and coagulation systems involves an organized proteolytic cascade, and the overactivation of these systems is a central pathogenic mechanism in several diseases. This review describes the role of complement and coagulation system activation in critical illness, particularly sepsis. The complexities of sepsis reveal significant knowledge gaps that can be compared to a profound abyss, highlighting the urgent need for further investigation and exploration. It is well recognized that the inflammatory network, coagulation, and complement systems are integral mechanisms through which multiple factors contribute to increased susceptibility to infection and may result in a disordered immune response during septic events in patients. Given the overlapping pathogenic mechanisms in sepsis, immunomodulatory therapies currently under development may be particularly beneficial for patients with sepsis who have concurrent infections. Herein, we present recent findings regarding the molecular relationships between the coagulation and complement pathways in the advancement of sepsis, and propose potential intervention targets related to the crosstalk between coagulation and complement, aiming to provide more valuable treatment of sepsis.

Abstract Background Sporadic parathyroid adenoma (PA) is the most common cause of hyperparathyroidism, yet the mechanisms involved in its pathogenesis remain incompletely understood. Methods Surgically removed PA samples, along with normal parathyroid gland (PG) tissues that were incidentally dissected during total thyroidectomy, were analysed using single‐cell RNA‐sequencing with the 10× Genomics Chromium Droplet platform and Cell Ranger software. Gene set variation analysis was conducted to characterise hallmark pathway gene signatures, and single‐cell regulatory network inference and clustering were utilised to analyse transcription factor regulons. Immunohistochemistry and immunofluorescence were performed to validate cellular components of PA tissues. siRNA knockdown and gene overexpression, alongside quantitative polymerase chain reaction, Western blotting and cell proliferation assays, were conducted for functional investigations. Results There was a pervasive increase in gene transcription in PA cells (PACs) compared with PG cells. This is associated with high expression of histone‐lysine N‐methyltransferase 2A (KMT2A). High KMT2A levels potentially contribute to promoting PAC proliferation through upregulation of the proto‐oncogene CCND2, which is mediated by the transcription factors signal transducer and activator of transcription 3 (STAT3) and GATA binding protein 3 (GATA3). PA tissues are heavily infiltrated with myeloid cells, while fibroblasts, endothelial cells and macrophages in PA tissues are commonly enriched with proinflammatory gene signatures relative to their counterparts in PG tissues. Conclusions We revealed the previously underappreciated involvement of the KMT2A‒STAT3/GATA3‒CCND2 axis and chronic inflammation in the pathogenesis of PA. These findings underscore the therapeutic promise of KMT2A inhibition and anti‐inflammatory strategies, highlighting the need for future investigations to translate these molecular insights into practical applications. Highlights Single‐cell RNA‐sequencing reveals a transcriptome catalogue comparing sporadic parathyroid adenomas (PAs) with normal parathyroid glands. PA cells show a pervasive increase in gene expression linked to KMT2A upregulation. KMT2A‐mediated STAT3 and GATA3 upregulation is key to promoting PA cell proliferation via cyclin D2. PAs exhibit a proinflammatory microenvironment, suggesting a potential role of chronic inflammation in PA pathogenesis.

Abstract A large‐scale AC/DC hybrid power network has recently been constructed in China because of the fast development of high voltage direct current (HVDC) technology. The receiving AC system must have sufficient system strength to guarantee the safe operation of the DC side. The paradox is that raising the AC side system's strength will increase the short‐circuit current (SCC) level, endangering the safety of the system. In order to resolve the contradiction between the system strength of multi‐infeed DC receiving‐end power networks and SCC, this paper proposes an optimal transmission switching strategy. Firstly, the system strength is characterised by the multi‐infeed short‐circuit ratio (MISCR), followed by constructing a comprehensive constraint set that considers the MISCR and SCC simultaneously. Secondly, to optimize SCC with MISCR and N−1 safety constraints, a mixed integer linear programming model is developed. Finally, to enhance the computing performance for large power systems, a new connectivity constraint set and decision space reduction techniques are provided. The proposed method is tested for effectiveness using the IEEE 30‐bus and IEEE 118‐bus systems.

Abstract Introduction The metastases of lung cancer to bilateral choroids symmetrically and simultaneously are very rare. Almost all patients with choroid metastasis can be treated with external beam radiotherapy in order to increase quality of life and preserve vision. Material and Methods We documented a case and studied the effect of icotinib on choroidal metastases in bilateral eyes simultaneously from pulmonary adenocarcinoma. Results A 49‐year‐old Chinese man presented with bilateral vision losing simultaneously for 4 weeks, it was as an initial presentation in the clinical. The examinations with ophthalmofundoscopy, ultrasonography, and fluorescein angiography showed the lesions in bilateral choroids, two solitary juxtapapillary yellow‐white choroidal metastases inferior to the optic discs with bleeding. Positron emission tomography confirmed the choroidal metastases and further proved that it was from lung cancer with lymph nodes and multiple bone metastasis. The biopsy taken from the lung by bronchoscopy and needle biopsy from supraclavicular lymph nodes revealed the pulmonary adenocarcinoma with epithelial growth factor receptor mutation (exon 21). The patient was treated with oral icotinib (125 mg, three times a day, TID). Five days after starting icotinib therapy, the patient's visions were rapidly recovered. Two months after the treatment with icotinib, the choroidal metastases regressed to small lesions, and the visions were preserved to before. The lung tumor and other metastatic lesions were partly regressive. There was no evidence of recurrence for eye lesions at 15‐months follow‐up. After 17 months treating by icotinib, the patient presented headache and dizzy with multiple brain metastases determined by magnetic resonance imaging; however, the lesions of the choroidal metastases remained progressing‐free. Almonertinib with radiotherapy were used to treat the brain metastases, and he is surviving with progress‐free more than 2 years until now. Conclusion Bilateral choroidal metastases from lung cancer symmetrically are very rare. Icotinib following by almonertinib was an alternative therapy for choroidal metastasis from non‐small cell lung cancer with epithelial growth factor receptor mutation.

Abstract Microgrids (MGs) provide a potential solution to accommodating renewable and distributed energy resources (DERs). MGs and the networked form of MGs, that is, networked MGs or NMGs, have received significant attention in the past two decades. However, several details that need to be considered for the practical operations of MGs and NMGs are often neglected in the literature. First, there is a need for a step‐by‐step sequence of operations (SOO) that clearly defines the procedures for changing the operation modes of MGs and NMGs for their reliable and resilient operation. Second, there is a need to develop a new control strategy for power exchange between networked MGs to ensure the information privacy and respect the electrical boundary of each MG. Third, there is a need for developing the model of MGs and NMGs in a real‐time simulator to safely evaluate the performance of the microgrid master controller (MMC) in making real‐time decisions and following the SOO. This paper proposes solutions for these three requirements and presents results to evaluate the performance of the proposed solutions. The IIT Campus Microgrid (ICM) and Bronzeville Community Microgrid (BCM), as well as their networked form, are selected as the practical testbeds and are modelled in real‐time digital simulator (RTDS). The RTDS model is interfaced with the MMC for real‐time data exchange and the operation of the MMC is tested to illustrate the importance of SOO in the real‐time control of MGs and NMGs.

Abstract Objective This study mainly observes changes in perioperative mineral bone metabolism‐related indicators and inflammatory factors in patients with secondary hyperparathyroidism (SHPT), and analyzed the correlation between mineral bone metabolism‐related indicators and inflammatory factors. Methods Clinical data were collected. The study detects mineral bone metabolism‐related indicators and inflammatory factor of perioperative patients with SHPT before and 4 days after operation. The production of high‐sensitivity c‐reactive protein (hs‐CRP) in human hepatocytes cells (LO2 cells) stimulated by different concentrations of parathyroid hormone‐associated protein was detected by enzyme‐linked immunosorbent assay, reverse‐transcription polymerase chain reaction (RT‐PCR), and western blot. Results The levels of mineral bone metabolism‐related indicators and hs‐CRP in SHPT group were significantly higher than those of control group. After operation, serum calcium, serum phosphorus, iPTH, FGF‐23 decreased, and the level of osteoblast active biomarkers increased, while the level of osteoclast active biomarkers decreased. The levels of hs‐CRP decreased significantly after operation. With the increase of PTHrP concentration, hs‐CRP level in supernatant of LO2 cells decreased first and then increased. RT‐PCR and western blot shows the same trend. Conclusion Parathyroidectomy can significantly improve bone resorption and inflammation in SHPT patients. We speculate that there may be an optimal range of PTH concentrations to minimize inflammation in the body.

Abstract The study of atomically dispersed metal‐nitrogen electrocatalysts is still limited in terms of understanding their catalytic mechanism because of the inability to precisely regulate the coordination number and type of N in combination with the metal elements. Inspired by the high catalytic activity and selectivity of natural enzymes, herein, we have designed and fabricated ultrathin carbon nanosheet‐supported Mn single‐atom catalysts (SACs) with a precise pyrrole‐type Mn‐N4 (PT‐MnN4) configuration using a bio‐mimicking strategy. The PT‐MnN4 SACs display outstanding oxygen reduction reaction (ORR) activity, with a half‐wave potential (E1/2) of 0.88 V (vs. revisible hydrogen electrode [RHE]) and extremely high stability in alkaline media. Moreover, superior ORR activities are also obtained, E1/2 of 0.73 V and 0.63 V in acid and neutral electrolytes, respectively, indicating the efficient pH‐universal ORR performances. The assembled zinc–air battery using the PT‐MnN4 SACs as air cathodes exhibits a high peak power density (175 mW cm−2) and long‐term stability up to 150 h, implying its promising application in metal–air batteries. This study has paved the way toward the rational design and precise regulation of single‐atom electrocatalysts.

Abstract Aim This study aimed to describe and visualize the current research state and collaborative networks in home care safety for older adults over the past 11 years to analyse the gaps of research and future research trends. Background The amount of research on safety in home care for older adults is increasing. It is necessary to understand the status of development and main research topics and identify the main contributors and their relationships. Methods A total of 2,631 publications were retrieved from the Web of Science. The external characteristics of the publications were summarized with the Web of Science and Histcite. Collaborative networks and keywords were analysed and visually displayed using analysis tools. Results The number of articles increased over the years. Articles were identified from 79 countries, 3,630 institutions, 647 journals and 11,691 authors, and complex cooperative relations among them and five research topics were identified. Conclusion Research on home care safety for older adults is developing steadily, and this field may be understood to a greater extent in the future. Countries, institutions and scholars need to cooperate more in this research field. Implications for Nursing Management This study contributes important information for understanding achievements in the research field of home care safety and provides insights into future research.

Abstract No nomogram models addressing the personalized prognosis evaluation of patients with gingival squamous cell carcinoma (GSCC) have been documented. We sought to establish nomograms to forecast overall survival (OS) and cancer‐specific survival (CSS) of patients with GSCC. We collected the detailed clinicopathological information of 2505 patients with GSCC from the Surveillance, Epidemiology and End Results (SEER) program. Afterward, we divided the 2505 cases into a modeling group (n = 1253) and an external validation cohort (n = 1252) via random split‐sample method. We developed the nomograms on the basis of the Kaplan‐Meier and multivariate Cox survival analysis of the modeling group and then split the modeling cohort into two parts based on cut‐off values: high‐ and low‐risk cohorts. An improved survival was shown in the low‐risk group compared to their counterpart, with a significant difference after the log‐rank test. The performance of the nomograms was evaluated via concordance‐index (C‐index), the area under the receiver operating characteristic curve (AUC), and calibration curves. All the C‐indexes and AUCs were greater than 0.7, showing high accuracy. Moreover, the calibrations showed that the actual observations were close to the 45° perfect reference line. In conclusion, we successfully developed two nomograms to provide individualized, patient‐specific estimates of OS and CSS available for risk‐stratification.

Abstract Background Alpha‐fetoprotein (AFP), routinely used for diagnosis of hepatocellular carcinoma (HCC), is limited with relatively low sensitivity and high false positivity in HBV‐related HCC (HBV‐HCC). Thus, an alternative approach was explored to improve specificity/sensitivity for diagnosis of HBV‐HCC, using the combination of AFP, inflammatory score, and liver function. Methods Chronic hepatitis B (CHB) (n = 510) and HBV‐HCC (n = 473) patients were identified retrospectively for this study. The diagnostic value of single vs combined biomarkers for HBV‐HCC was analyzed, using ROC curve. Results It was observed that elderliness, male sex, cirrhosis, HBeAg+ or no‐antiviral therapy, and elevation of ALT, AST, neutrophil‐lymphocyte ratio (NLR), and AFP were associated with developing HBV‐HCC. However, the cut‐off ALT defined by Chinese standard, but not by AASLD, was a risk factor. Interestingly, AFP of HBeAg‐ HBV‐HCC patients without cirrhosis was significantly higher than that of the HBeAg+ patients. AUC values for AFP, ALT, AST, or NLR were 0.84 (95% CI: 0.815‐0.862), 0.533 (95% CI: 0.501‐0.565), 0.696 (95% CI: 0.666‐0.725), or 0.684 (95% CI: 0.654‐0.713) with optimal cut‐off at 7.21 ng/mL, 43 IU/mL, 38 IU/mL, or 2.61, respectively. Combination of AFP with ALT, AST, and NLR improved the diagnostic performance for HBV‐HCC, compared to any of the single biomarkers or any other combinations among these patients (except no‐cirrhosis). Conclusions Elderliness, male sex, elevated ALT, AST, NLR, AFP, cirrhosis, HBeAg+, and no‐antiviral treatment were independent risk factors for HBV‐HCC. AASLD standard of ALT cut‐off value may not be suitable for the Chinese population. Regular monitoring of HCC among HBeAg‐ patients with abnormal AFP may improve the management of HBV‐HCC. The diagnostic performance of AFP combined with ALT, AST, and NLR for HBV‐HCC was superior to single biomarker or any other combinations among these patients, and its diagnostic equation can be used as useful tool for differentiation of HBV‐HCC from CHB.

Abstract Thermal burn is a complex injury that induces pronounced inflammatory reactions and destruction of the microvasculature. Lidocaine, which is broadly used for pain relief, has been reported have the capacity to modulate wound healing processes. Seventeen burn injury patients with no treatment and nine controls were included in this study. The expression of 15 candidate miRNAs in the dermis samples were detected by qRT‐PCR. The target genes were predicted using online bioinformatics tools and confirmed by dual luciferase assay and immunoblotting. The function of miR‐486 and miR‐663 on skin fibroblasts keratinocytes were determined by MTT assay and flow cytometry. The level of miR‐486 and miR‐663 was increased after burn injury, meanwhile, the highest level of miR‐486 and miR‐663 was found after lidocaine treatment. We identified that BCL2L14 was a direct target of both miR‐486 and miR‐663. Meanwhile, overexpression of miR‐486 or miR‐663 inhibit apoptosis and promoted proliferation of human skin fibroblasts keratinocytes. These results indicated that lidocaine treatment promoted the skin healing after thermal injury through up‐regulating miR‐486 and miR‐663 expression, and partially explained how lidocaine modulates wound healing processes. This may provide an evidence for the therapeutic effect of lidocaine during skin healing process. However, the underlying mechanisms need to be further unveiled.

Abstract Understanding hot carrier dynamics between plasmonic nanomaterials and its adsorbate is of great importance for plasmon‐enhanced photoelectronic processes such as photocatalysis, optical sensing and spectroscopic analysis. However, it is often challenging to identify specific dominant mechanisms for a given process because of the complex pathways and ultrafast interactive dynamics of the photoelectrons. Here, using CO2 reduction as an example, the underlying mechanisms of plasmon‐driven catalysis at the single‐molecule level using time‐dependent density functional theory calculations is clearly probed. The CO2 molecule adsorbed on two typical nanoclusters, Ag20 and Ag147, is photoreduced by optically excited plasmon, accompanied by the excitation of asymmetric stretching and bending modes of CO2. A nonlinear relationship has been identified between laser intensity and reaction rate, demonstrating a synergic interplay and transition from indirect hot‐electron transfer to direct charge transfer, enacted by strong localized surface plasmons. These findings offer new insights for CO2 photoreduction and for the design of effective pathways toward highly efficient plasmon‐mediated photocatalysis.

Abstract Due to the complex geostress and mining conditions in the coal seam with depth of 800 m, stability of surrounding rock for gob‐side entry retaining is very difficult to achieve. In this paper, we firstly propose an innovative bolt‐grouting controlled roof‐cutting for gob‐side entry retaining (BCR‐GER) approach for deep coal mines. Secondly, a mechanical model of “surrounding rock‐supporting body” for BCR‐GER is constructed, which consists of coal wall, roadside props, and gangues in gob (the whole supporting body). Thirdly, the key parameters (ie, cutting height, cutting angle, grouting cable length, and row of roadside props) are designed. Finally, field practice was applied at the No. 31120 haulage roadway of the Suncun coal mine in China, and in situ investigations were conducted for verification. Field measurement results show that maximum convergences of roof‐to‐floor and side‐to‐side were 264 mm and 113 mm, respectively. What is more, the maximum support resistance of roadside props was reduced by approximately 58%. The deformation and failure of surrounding rock were effectively controlled, and the pressure on roadside props was greatly reduced. This research fully considers the bearing properties of gangues in gob, eliminates the secondary disasters caused by borehole blasting, and provides guidance and reference for deep surrounding rock control of the same or similar gob‐side entry.

Abstract Natural and non‐natural factors have combined effects on the trajectory of COVID‐19 pandemic, but it is difficult to make them separate. To address this problem, a two‐stepped methodology is proposed. First, a compound natural factor (CNF) model is developed via assigning weight to each of seven investigated natural factors, that is temperature, humidity, visibility, wind speed, barometric pressure, aerosol, and vegetation in order to show their coupling relationship with the COVID‐19 trajectory. Onward, the empirical distribution based framework (EDBF) is employed to iteratively optimize the coupling relationship between trajectory and CNF to express the real interaction. In addition, the collected data is considered from the backdate, that is about 23 days—which contains 14‐days incubation period and 9‐days invalid human response time—due to the nonavailability of prior information about the natural spreading of virus without any human intervention(s), and also lag effects of the weather change and social interventions on the observed trajectory due to the COVID‐19 incubation period; Second, the optimized CNF‐plus‐polynomial model is used to predict the future trajectory of COVID‐19. Results revealed that aerosol and visibility show the higher contribution to transmission, wind speed to death, and humidity followed by barometric pressure dominate the recovery rates, respectively. Consequently, the average effect of environmental change to COVID‐19 trajectory in China is minor in all variables, that is about −0.3%, +0.3%, and +0.1%, respectively. In this research, the response analysis of COVID‐19 trajectory to the compound natural interactions presents a new prospect on the part of global pandemic trajectory to environmental changes.

Diesel engines are widely used in railway systems, particularly in freight trains. Despite their high efficiency in energy conversion, they usually generate high levels of acoustics pollution during operation. In order to mitigate this problem, a series of active/passive acoustics control methods are used to reduce noise. Most of these methods are only effective if the prior knowledge of sources is given. In other words, it is essential to recognize the acoustics source. Variational mode decomposition (VMD) is a signal processing method that enhances the signal corrupted by background noise. However, the decomposed results of VMD depend on their mode parameter and penalty parameter. Therefore, an evaluation method based on system modal parameters (natural frequency and damping ratio) is proposed to select the mode parameter, and the penalty parameter can be selected from the power spectra of signals. In order to increase the accuracy of decomposition for diesel engines and find out the sources of acoustics, a method combining VMD, fast independent component analysis, and Hilbert transformation (VMD-FastICA-HT) is proposed for the separation and identification of different sources for diesel engines. The optimization results indicate that when the penalty parameter value is 1.5 to 16 times the maximum signal amplitude, better decomposition results can be achieved. Therefore, the separated independent acoustics are more accurate in source identification. Furthermore, both simulation data and in situ operational data of diesel engines for vehicles are used to validate the effectiveness of the proposed method.

Abstract In the mammalian ovaries, dormant primordial follicles represent the reproductive reserve of individual females. Recently, stimulating the activation of primordial follicles in vitro has been practiced, making the utilization of those dormant follicles to treat female infertility possible. However, there are still lacks of effective upstream molecule and strategy to elevate follicle activation in vivo. In the current study, we revealed that growth factor EGF improved a transiently primordial follicle activation in mice by elevating the CDC42‐PI3K signaling activity, and EGF treatment also improved the activation and development of human follicles in ovarian cortical pieces. Using a liquid‐solid phase transition bio‐gel as a carrier, an efficient in vivo activation system was established by ovarian topical EGF administration to living mice. We found that EGF treatment led to an increase of primordial follicle activation in short time but had no effect on long‐term fertility in females. By establishing an inducible premature ovarian insufficiency (POI) mouse model through selectively ablating growing follicles in Zp3‐Cre;iDTR mice, we further revealed that our in vivo EGF treatment system improved primordial follicle activation and ovulation of POI ovaries significantly. Taken together, our results revealed that in situ ovarian EGF administration could improve the activation of primordial follicles in living animals, and manipulating activation and development of primordial follicles in vivo might be an efficient approach to improve reproductive health in women.

Human gait phase recognition is a significant technology for rehabilitation training robot, human disease diagnosis, artificial prosthesis, and so on. The efficient design of the recognition method for gait information is the key issue in the current gait phase division and eigenvalues extraction research. In this paper, a novel voting-weighted integrated neural network (VWI-DNN) is proposed to detect different gait phases from multidimensional acceleration signals. More specifically, it first employs a gait information acquisition system to collect different IMU sensors data fixed on the human lower limb. Then, with dimensionality reduction and four-phase division preprocessing, key features are selected and merged as unified vectors to learn common and domain knowledge in time domain. Next, multiple refined DNNs are transferred to design a multistream integrated neural network, which utilizes the mixture-granularity information to exploit high-dimensional feature representative. Finally, a voting-weighted function is developed to fuse different submodels as a unified representation for distinguishing small discrepancy among different gait phases. The end-to-end implementation of the VWI-DNN model is fine-tuned by the loss optimization of gradient back-propagation. Experimental results demonstrate the outperforming performance of the proposed method with higher classification accuracy compared with the other methods, of which classification accuracy and macro-F1 is up to 99.5%. More discussions are provided to indicate the potential applications in combination with other works.