Deep learning-based anatomical position recognition for gastroscopic examination.

BACKGROUND: The gastroscopic examination is a preferred method for the detection of upper gastrointestinal lesions. However, gastroscopic examination has high requirements for doctors, especially for the strict position and quantity of the archived images. These requirements are challenging for the education and training of junior doctors. OBJECTIVE: The purpose of this study is to use deep learning to develop automatic position recognition technology for gastroscopic examination. METHODS: A total of 17182 gastroscopic images in eight anatomical position categories are collected. Convolutional neural network model MogaNet is used to identify all the anatomical positions of the stomach for gastroscopic examination The performance of four models is evaluated by sensitivity, precision, and F1 score. RESULTS: The average sensitivity of the method proposed is 0.963, which is 0.074, 0.066 and 0.065 higher than ResNet, GoogleNet and SqueezeNet, respectively. The average precision of the method proposed is 0.964, which is 0.072, 0.067 and 0.068 higher than ResNet, GoogleNet, and SqueezeNet, respectively. And the average F1-Score of the method proposed is 0.964, which is 0.074, 0.067 and 0.067 higher than ResNet, GoogleNet, and SqueezeNet, respectively. The results of the t-test show that the method proposed is significantly different from other methods (p< 0.05). CONCLUSION: The method proposed exhibits the best performance for anatomical positions recognition. And the method proposed can help junior doctors meet the requirements of completeness of gastroscopic examination and the number and position of archived images quickly.

Combination of high-frequency electrocautery therapy and ALA-PDT in hyperkeratotic vulvar lichen sclerosus: Series of seven cases.

SIGNIFICANCE: ALA-PDT effectively treats Vulvar lichen sclerosus et atrophicus (VLSA), but it requires multiple repetitions for satisfactory results. To enhance efficacy, we employed a combination of high-frequency electrocautery therapy and ALA-PDT in treating seven VLSA patients. APPROACH: Lesions and leukoplakia in the seven women with VLSA were removed using a high-frequency generator. PDT was administered after wound healing, and it was repeated six times. Follow-up assessments were carried out at 1, 3, and 6 months to evaluate the severity of pruritus and investigate lesion repigmentation. RESULTS: Following the combined therapy, the disappearance of pruritus was observed in all patients, and normal color and thickness were restored to their skin. Two patients reported mild pruritus with a score of 2 one month after treatment, which persisted until the 6-month follow-up, while the remaining patients remained free from pruritus. No recurrence of skin lesions was observed in any of the patients. CONCLUSIONS: The combined therapy for the treatment of VLSA is found to be convenient, effective, and easily promotable.

Prediction of outpatient waiting time: using machine learning in a tertiary children's hospital.

Background: Accurately predicting waiting time for patients is crucial for effective hospital management. The present study examined the prediction of outpatient waiting time in a Chinese pediatric hospital through the use of machine learning algorithms. If patients are informed about their waiting time in advance, they can make more informed decisions and better plan their visit on the day of admission. Methods: First, a novel classification method for the outpatient clinic in the Chinese pediatric hospital was proposed, which was based on medical knowledge and statistical analysis. Subsequently, four machine learning algorithms [linear regression (LR), random forest (RF), gradient boosting decision tree (GBDT), and K-nearest neighbor (KNN)] were used to construct prediction models of the waiting time of patients in four department categories. Results: The three machine learning algorithms outperformed LR in the four department categories. The optimal model for Internal Medicine Department I was the RF model, with a mean absolute error (MAE) of 5.03 minutes, which was 47.60% lower than that of the LR model. The optimal model for the other three categories was the GBDT model. The MAE of the GBDT model was decreased by 28.26%, 35.86%, and 33.10%, respectively compared to that of the LR model. Conclusions: Machine learning can predict the outpatient waiting time of pediatric hospitals well and ease patient anxiety when waiting in line without medical appointments. This study offers key insights into enhancing healthcare services and reaffirms the dedication of Chinese pediatric hospitals to providing efficient and patient-centric care.

Wireless and Closed-Loop Smart Dressing for Exudate Management and On-Demand Treatment of Chronic Wounds.

Chronic wounds have become a significant threat to people's physical and mental health and have increased the burden of social medical care. Intelligent wound dressing (IWD) with wound condition monitoring and closed-loop on-demand drug therapy can shorten the healing process and alleviate patient suffering. However, single-function wound dressings cannot meet the current needs of chronic wound treatment. Here, a wearable IWD consisting of wound exudate management, sensor monitoring, closed-loop therapy, and flexible circuit modules is reported, which can achieve effective synergy between wound exudate management and on-demand wound therapy. The dressing is attached to the wound site, and the wound exudate is spontaneously pumped into the microfluidic channel for storage. Meanwhile, the IWD can detect the state of the wound through the temperature and humidity sensor, and use this as feedback to control the liquid metal (LM) heater through a smartphone, thereby realizing the on-demand drug release from the hydrogel. In a mouse model of infected wounds, IWD accelerates wound healing by reducing inflammatory responses, promoting angiogenesis and collagen deposition.

The latest advances in high content screening in microfluidic devices.

INTRODUCTION: High content screening (HCS) is an important tool for drug screening. However, the potential of HCS in the field of drug screening and synthetic biology is limited by traditional culture platforms that use multi-well plates, which have several disadvantages. Recently, microfluidic devices have gradually been applied in HCS, which significantly reduces experimental costs, increases assay throughput, and improves the accuracy of drug screening. AREAS COVERED: This review provides an overview of microfluidic devices for high-content screening in drug discovery platforms, including droplet, microarray, and organs-on-chip technologies. EXPERT OPINION: HCS is a promising technology increasingly adopted by the pharmaceutical industry as well as academic researchers for drug discovery and screening. In particular, microfluidic-based HCS shows unique advantages, and microfluidics technology has promoted significant advancements and broader usage and applicability of HCS in drug discovery. With the integration of stem cell, gene editing technology, and other biological technologies, microfluidics-based HCS will expand the application scope of personalized disease and drug screening models. The authors anticipate rapid developments in this field, with microfluidic-based approaches becoming increasingly important in HCS applications.

Portable general microfluidic device with complex electric field regulation functions for electrokinetic experiments.

Electrokinetic sample manipulation is a key step for many kinds of microfluidic chips to achieve various functions, such as particle focusing and separation, fluid pumping and material synthesis. But these microfluidic experiments usually rely on large-scale signal generators for power supply, microscopes for imaging and other instruments for analysis, which hampers the portable process of microfluidic technology. Inspired by this situation, we herein designed a portable general microfluidic device (PGMD) with complex electric field regulation functions, which can accurately regulate static or continuous fluid samples. Through the graphical user interface (GUI) and modular design, the PGMD can generate multiple different electrical signals, and the micro-flow of fluid can be pumped through the built-in micropump, which can meet the requirements of most microfluidic experiments. Photos or videos of the microfluidic chip captured by the built-in microscope are received and displayed by a smartphone. We carried out a variety of microfluidic experiments such as induced-charge electroosmosis (ICEO), particle beam exit switching, thermal buoyancy flow and dielectrophoresis (DEP) on the PGMD. In addition, the PGMD can perform rapid microalgae concentration estimation in an outdoor environment, which can be used to guide microalgae cultivation, further demonstrating the development potential of this device in the field of microbial applications. Numerous results show that the PGMD has a high degree of integration and strong reliability, which expands the application of microfluidic electrokinetic experiments and provides technical support for the integration and portability of microfluidic experimental devices.

Liquid metal droplet motion transferred from an alkaline solution by a robot arm.

The excellent motion performance of gallium-based liquid metals (LMs) upon the application of a modest electric field has provided a new opportunity for the development of autonomous soft robots. However, the locomotion of LMs often appears in an alkaline solution, which hampers the application under other different conditions. In this work, a novel robot arm is designed to transfer the motion of the LM from an alkaline solution in a synchronous drive mode. The liquid metal droplet (LMD) at the bottom of the robot arm is actuated using a DC voltage to provide the driving force for the system. By introducing an end effector at the center of the robot arm, the synchronous motion of the system is replicated and can be applied to different situations. The theoretical understanding of continuous electrowetting (CEW) at the LM interface is explained, and then the motion performance of the robot arm against the function of the applied voltage and driving direction is investigated. Moreover, several applications using this robot arm, such as pattern drawing, cargo transportation, and drug concentration detection, are demonstrated. The presented robot arm has the potential to observably expand the application fields of the LM.

Hypertensive vascular and cardiac remodeling protection by allicin in spontaneous hypertension rats via CaMK â ¡/NF-κB pathway.

Allicin is the main active component of Traditional Chinese medicine, garlic. It is widely used to treat cardiovascular diseases. Our previous studies have confirmed that allicin significantly reduces blood pressure in Spontaneous Hypertension Rats (SHRs). However, the reports studying the effect of allicin on vascular and cardiac remodeling caused by hypertension are few, with their underlying mechanism not being studied in detail or fully elucidated. In this study, we treated 12-week-old SHRs with allicin for 4 weeks. After 4 weeks, allicin was shown to improve vascular and cardiac remodeling in SHRs, as evidenced by reduced cardiac left ventricular wall thickness, aortic vessel thickness, and reduced proliferating cell nuclear antigen (PCNA) and smooth muscle actin (α-SMA), and increased expression of and smooth muscle 22α (SM 22α). Additionally, allicin reduced serum IL-1ß, IL-6, and TNF-α levels, improved calcium homeostasis in cardiomyocytes, downregulated calcium transportation-related CaMK II and inflammation-related NF-κB and NLRP3, which were observed in smooth muscle cells and cardiomyocytes. Thus, we inferred that allicin protected hypertensive vascular and cardiac remodeling in Spontaneous Hypertensive Rats by inhibiting the activation of the CaMK II/ NF-κB pathway. This study also provided new mechanistic insights into the anti-hypertensive vascular and cardiac remodeling effects of allicin, highlighting its therapeutic potential.

Numerical investigation of field-effect control on hybrid electrokinetics for continuous and position-tunable nanoparticle concentration in microfluidics.

We introduce herein an effective way for continuous delivery and position-switchable trapping of nanoparticles via field-effect control on hybrid electrokinetics (HEK). Flow field-effect transistor exploiting HEK delicately combines horizontal linear electroosmosis and transversal nonlinear electroosmosis of a shiftable flow stagnation line (FSL) on gate terminals under DC-biased AC forcing. The microfluidic nanoparticle concentrator proposed herein makes use of a simple device geometry, in which an individual or a series of planar metal strips serving as gate electrode (GE) are subjected to a hybrid gate voltage signal and arranged in parallel between a pair of 3D driving electrodes. On the application of a DC-biased AC electric field across channel length direction, all the GE are electrochemically polarized, and the action of imposed hybrid electric field on the multiple-frequency bipolar counterions within the composite-induced double layer generates two counter-rotating induced-charge electroosmotic (ICEO) micro-vortices on top of each GE. Symmetry breaking in ICEO flow profile occurs once the gate voltage deviates from natural floating potential of corresponding GE. The gate voltage offset not only results in an additional pump motion of working fluid for enhanced electroosmotic transport but also directly changes the location of FSL where nanoparticles are preferentially collected by field-effect HEK. Our results of field-effect control on HEK are supposed to guide an elaborate design of flexible electrokinetic frameworks embedding coplanar metal strips for a high degree of freedom analyte manipulation in modern micro-total-analytical systems.

Numerical characterization of inter-core coalescence by AC dielectrophoresis in double-emulsion droplets.

The utilization of an alternating current electric field provides a good means to achieve controlled coalescence between paired inner cores encapsulated in water-in-oil-in-water double-emulsion (DE) droplets. Although previous studies have experimentally determined the conditions under which inter-core electrokinetic fusion occurs, the transient interfacial dielectrophoretic (DEP) dynamics key to understand the underlying fluid mechanics is still unclear from a physical point of view. By coupling DEP motion of two-phase flow to phase-field formulation, bulk-coupled numerical simulations are conducted to characterize the spatial-temporal evolution of the surface charge wave and the resulting nonlinear electrical force induced at both the core/shell and medium/shell oil/water interfaces. The effect of interfacial charge relaxation and droplet geometry on inter-core attractive dipolar interaction is investigated within a wide parametric space, and four distinct device operation modes, including normal inter-core fusion, shell elongation, partial core leakage, and complete core release, are well distinguished from one another by flow regime argumentation. Our results herein reveal for the first time the hitherto unknown transient electrohydrodynamic fluid motion of DE droplet driven by Maxwell-Wagner structural polarization. The dynamic simulation method proposed in present study points out an effective outlet to predict the nonlinear electrokinetic behavior of multicore DE droplets for realizing a more controlled triggering of microscale reactions for a wide range of applications in drug discovery, skin care, and food industry.

Deep convolutional neural networks for regular texture recognition.

Regular textures are frequently found in man-made environments and some biological and physical images. There are a wide range of applications for recognizing and locating regular textures. In this work, we used deep convolutional neural networks (CNNs) as a general method for modelling and classifying regular and irregular textures. We created a new regular texture database and investigated two sets of deep CNNs-based methods for regular and irregular texture classification. First, the classic CNN models (e.g. inception, residual network, etc.) were used in a standard way. These two-class CNN classifiers were trained by fine-tuning networks using our new regular texture database. Next, we transformed the trained filter features of the last convolutional layer into a vector representation using Fisher Vector pooling (FV). Such representations can be efficiently used for a wide range of machine learning tasks such as classification or clustering, thus more transferable from one domain to another. Our experiments show that the standard CNNs attained sufficient accuracy for regular texture recognition tasks. The Fisher representations combined with support vector machine (SVM) also showed high performance for regular and irregular texture classification. We also find CNNs performs sub-optimally for long-range patterns, despite the fact that their fully-connected layers pool local features into a global image representation.

A visual portable microfluidic experimental device with multiple electric field regulation functions.

High portability and miniaturization are two of the most important objectives pursued by microfluidic methods. However, there remain many challenges for the design of portable and visual microfluidic devices (e.g., electrokinetic experiments) due to the use of a microscope and power supply. To this end, we report a visual portable microfluidic experimental device (PMED) with multiple electric field regulation functions, which can realize the electric field regulation functions of various basic microfluidic experiments through modular design. The internal reaction process of the microfluidic chip is displayed by a smartphone, and the experimental results are analyzed using a mobile phone application (APP). Taking the induced-charge electroosmosis (ICEO) particle focusing phenomenon as an example, we carried out detailed experiments on PMED and obtained conclusions consistent with numerical simulations. In addition to ICEO experiments, other functions such as alternating electroosmosis (ACEO), thermal buoyancy convection, and dielectrophoresis (DEP) can be realized by replacing module-specific covers. The device expands the application of microfluidic experiments and provides a certain reference for the further integration and portability of subsequent microfluidic experiment devices.

Fluid pumping by liquid metal droplet utilizing ac electric field.

We report a unique phenomenon in which liquid metal droplets (LMDs) under a pure ac electric field pump fluid. Unlike the directional pumping that occurs upon reversing the electric field polarity under a dc signal, this phenomenon allows the direction of fluid motion to be switched by simply shifting the position of the LMD within the cylindrical chamber. The physical mechanism behind this phenomenon has been termed Marangoni flow, caused by nonlinear electrocapillary stress. Under the influence of a localized, asymmetric ac electric field, the polarizable surface of the position-offset LMD produces a net time-averaged interfacial tension gradient that scales with twice the field strength, and thus pumps fluid unidirectionally. However, the traditional linear RC circuit polarization model of the LMD/electrolyte interface fails to capture the correct pump-flow direction when the thickness of the LMD oxide skin is non-negligible compared to the Debye length. Therefore, we developed a physical description by treating the oxide layer as a distributed capacitance with variable thickness and connected with the electric double layer. The flow profile is visualized via microparticle imaging velocimetry, and excellent consistency is found with simulation results obtained from the proposed nonlinear model. Furthermore, we investigate the effects of relevant parameters on fluid pumping and discuss a special phenomenon that does not exist in dc control systems. To our knowledge, no previous work addresses LMDs in this manner and uses a zero-mean ac electric field to achieve stable, adjustable directional pumping of a low-conductivity solution.

Revealing Calcium Signaling Pathway as Novel Mechanism of Danhong Injection for Treating Acute Myocardial Infarction by Systems Pharmacology and Experiment Validation.

Introduction: Danhong injection (DHI) is a traditional Chinese medicine preparation commonly used in the clinical treatment of acute myocardial infarction (AMI). In this study, the active components of DHI and its mechanism in the treatment of AMI were investigated. Methods: The chemical components of DHI were detected by the ultra-high-performance liquid chromatography-linear trap quadrupole-orbitrap-tandem mass spectrometry (UHPLC-LTQ-Orbitrap-MS/MS), and the targets and pathways of DHI in the treatment of AMI were analyzed by systems pharmacology, which was verified by molecular docking and animal experiments. Results: A total of 12 active components of DHI were obtained, and 158 common targets of component and disease were identified by systems pharmacology. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis results showed that DHI is closely related to the calcium signaling pathway in the treatment of AMI. Molecular docking showed that the key target protein has good binding affinity to related compounds. The experimental results showed that compared with the model group, LVAWs, EF, and FS significantly (p < 0.05) increased in the DHI group. The percentage of myocardial infarction significantly (p < 0.01) decreased, both in the ventricular and total cardiac regions, and the pathological damage of myocardial tissue also decreased. In addition, the expression of the protein CaMK II decreased (p < 0.01) and the expression of SERCA significantly increased (p < 0.01). Conclusion: This study revealed that ferulic acid, caffeic acid and rosmarinic acid could inhibit AMI by regulating PLB, CaMK II, SERCA, etc. And mechanistically, calcium signaling pathway was critically involved. Combination of systems pharmacology prediction with experimental validation may provide a scientific basis for in-depth clinical investigation of the material basis of DHI.

Desktop-level small automatic guided vehicle driven by a liquid metal droplet.

Gallium-based liquid metals (LMs) are a new type of intelligent material, and their ability to move under the action of an electric field provides new opportunities for the design of small flexible vehicles. However, due to the extremely high fluidity of LMs and the poor automatic control ability of LM vehicles, it's still a huge challenge to control the movement of LMs flexibly and accurately. Therefore, in this paper, a small traction vehicle is designed by putting the flexible LM in rigid armor to make the movement more controllable. Moreover, a desktop-level small automatic guided vehicle (sAGV) system is built by using an external control circuit to follow a predetermined trajectory. Firstly, the basic characteristics of the vehicles driven by a LM droplet are simulated and analyzed. Then the effects of different factors on the movement velocity of the vehicles are measured by experiment. Finally, as a preliminary application test, the sAGV system is used to control the vehicles following a specific trajectory and realize the targeted transportation of cargos. The sAGV system designed in this paper can realize the automatic and precise control of the movement of the small vehicle. The current findings will inspire the further construction of complex small operating systems and the realization of accurate control.

DC electric field-driven heartbeat phenomenon of gallium-based liquid metal on a floating electrode.

The heart beating phenomenon of room temperature liquid metal (LM) mercury has attracted much attention in the past years, but its research and application are limited because of the low vapor pressure and high toxicity. Here, a fundamental scientific finding is reported that the non-toxic eutectic gallium indium (EGaIn) alloy droplets beat periodically at a certain frequency based on a floating electrode under the stimulation of the direct current (DC) field. The essential characteristics of heart beating are the displacement and the projected area change of the LM droplet. The mechanism of this phenomenon is the self-regulation of interfacial tension caused by chemical oxidation, chemical corrosion, and continuous electrowetting. In this article, a series of experiments are also carried out to examine the effects of different factors on the heartbeat, such as voltage, the volume of the droplet, the droplet immersion depth, the electrolyte solution concentration, the distance of electrodes, and the type of floating electrode. Finally, the heartbeat state and application boundary of the LM droplet under different conditions are summarized by imitating the human life process. The periodic changes of the LM droplet under an external DC electric field provide a new method to simulate the beating of the heart artificially, and can be applied to the research of organ chip fluid pumping in the future.

Process of Glucose Increases Rather Than Constant High Glucose Was the Main Cause of Abnormal Glucose Induced Glomerulus Epithelial Cells Inflammatory Response.

Abnormal glycemia is frequently along with nephritis, whose pathogenesis is unexplicit. Here, we investigated the effects of abnormal glucose on the renal glomerulus epithelial cells by stimulating immortalized bovine renal glomerulus epithelial (MDBK) cells with five different levels of glucose, including low glucose (2.5 mM for 48 h, LG), normal glucose (5 mM for 48 h, NG), high glucose (25 mM for 48 h, HG), increasing glucose (24 h of 2.5 mM glucose followed by 24 h of 25 mM, IG), and reducing glucose (24 h of 25 mM glucose followed by 24 h of 2.5 mM, RG). The results showed that LG and RG treatments had nonsignificant effects (p > 0.05) on the viability of MDBK cells. HG treatment decreased the viabilities of cells (p < 0.01) without triggering an apparent inflammatory response by activating the nox4/ROS/p53/caspase-3-mediated apoptosis pathway. IG treatment decreased the viabilities of cells significantly (p < 0.01) with high levels of pro-inflammatory cytokines IL-1ß and IL-18 in the supernatant (p < 0.05) by triggering the txnip/nlrp3/gsdmd-mediated pyroptosis pathway. These results indicated that the process of glucose increase rather than the constant high glucose was the main cause of abnormal glucose-induced MDBK cell inflammatory death, prompting that the process of glycemia increases might be mainly responsible for the nephritis in diabetic nephropathy, underlining the importance of glycemic control in diabetes patients.

Secukinumab-induced multiple lentigines in areas of resolved psoriatic plaques: A case report and literature review.

Psoriasis is a systemic inflammatory disease commonly associated with postinflammatory hyper- and hypo-pigmentation. Psoriasis-related cytokines such as IL-17 and TNF can contribute to these pigmentation changes by regulating both the growth and pigment production of melanocytes. Here, we present the first reported the case of a patient with a 10-year history of severe psoriasis vulgaris, who developed multiple lentigines in areas of resolved psoriatic plaques during anti-IL-17A antibody secukinumab.

Small universal mechanical module driven by a liquid metal droplet.

Gallium-based liquid metal droplets (LMDs) from micro-electromechanical systems (MEMS) have gained much attention due to their precise and sensitive controllability under an electric field. Considerable research progress has been made in the field of actuators by taking advantage of the continuous electrowetting (CEW) present within the solution. However, the motion generated is confined within the specific liquid environment and is lacking a way to transmit its motion outwardly, which undoubtedly serves as the greatest obstacle restricting any further development. Therefore, a driving module is proposed to generate rotational motion outside the solution for universality. Its performance can be easily tuned by adjusting the applied voltage. As an example of further application, the module is designed in the form of a pump that realizes the continuous/intermittent propulsion to mimic the veins/arteries of the human body without the problem in the previous LMD-based pumps. The feasibility of this pump in the on-chip in vitro analysis is proved by preparing a dynamic cell culture to simulate the movement of biofluids within human bodies. This study proposes an optional solution with an LMD-based motor for generating rotational motion and to expand current research on soft materials in actuators.

Fenretinide regulates macrophage polarization to protect against experimental colitis induced by dextran sulfate sodium.

Fenretinide (4-HPR), a synthetic retinoid, has attracted attention for its anti-inflammation activity. However, few studies have evaluated the effects of 4-HPR on ulcerative colitis (UC). The present study was performed to investigate the therapeutic effects of 4-HPR on UC, and to explore the mechanisms mainly focused on macrophage polarization involved in this progress. Intraperitoneally administered 4-HPR particularly at dose of 100 mg/kg obviously alleviated UC symptoms and restrained the mRNA expression of colonic IL-1ß, IL-6, and TNF-α in dextran sulfate sodium (DSS)-induced mice. Further analysis showed that 4-HPR decreased the mRNA expression of M1 macrophage markers IL-12 and iNOS, while increased M2 macrophage markers Ym1, Arg1 and MRC1 in colonic tissue of mice received DSS. Consistently, an in vitro study revealed that 4-HPR decreased inflammatory response and M1 polarization, while enhanced M2 polarization in LPS-induced RAW264.7 cells. Interestingly, 4-HPR remarkably activated PPAR-γ which was an important regulator of macrophage polarization both in colonic tissue of UC mice and in LPS-induced RAW264.7 cells. Furthermore, these effects of 4-HPR in vivo and in vitro including anti-inflammation and modulation of macrophage polarization were partially abolished by treatment with PPAR-γ antagonist GW9662, indicating that 4-HPR activated PPAR-γ to exert its activities. Taken together, this study demonstrated that 4-HPR might be a potent anti-UC agent that works by regulating macrophage polarization via PPARγ.