Generative adversarial network-based deep learning approach in classification of retinal conditions with optical coherence tomography images.

PURPOSE: To determine whether a deep learning approach using generative adversarial networks (GANs) is beneficial for the classification of retinal conditions with Optical coherence tomography (OCT) images. METHODS: Our study utilized 84,452 retinal OCT images obtained from a publicly available dataset (Kermany Dataset). Employing GAN, synthetic OCT images are produced to balance classes of retinal disorders. A deep learning classification model is constructed using pretrained deep neural networks (DNNs), and outcomes are evaluated using 2082 images collected from patients who visited the Department of Ophthalmology and the Department of Endocrinology and Metabolism at the Tri-service General Hospital in Taipei from January 2017 to December 2021. RESULTS: The highest classification accuracies accomplished by deep learning machines trained on the unbalanced dataset for its training set, validation set, fivefold cross validation (CV), Kermany test set, and TSGH test set were 97.73%, 96.51%, 97.14%, 99.59%, and 81.03%, respectively. The highest classification accuracies accomplished by deep learning machines trained on the synthesis-balanced dataset for its training set, validation set, fivefold CV, Kermany test set, and TSGH test set were 98.60%, 98.41%, 98.52%, 99.38%, and 84.92%, respectively. In comparing the highest accuracies, deep learning machines trained on the synthesis-balanced dataset outperformed deep learning machines trained on the unbalanced dataset for the training set, validation set, fivefold CV, and TSGH test set. CONCLUSIONS: Overall, deep learning machines on a synthesis-balanced dataset demonstrated to be advantageous over deep learning machines trained on an unbalanced dataset for the classification of retinal conditions.

Using machine learning to determine the correlation between physiological and environmental parameters and the induction of acute mountain sickness.

BACKGROUND: Recent studies on acute mountain sickness (AMS) have used fixed-location and fixed-time measurements of environmental and physiological variable to determine the influence of AMS-associated factors in the human body. This study aims to measure, in real time, environmental conditions and physiological variables of participants in high-altitude regions to develop an AMS risk evaluation model to forecast prospective development of AMS so its onset can be prevented. RESULTS: Thirty-two participants were recruited, namely 25 men and 7 women, and they hiked from Cuifeng Mountain Forest Park parking lot (altitude: 2300 m) to Wuling (altitude: 3275 m). Regression and classification machine learning analyses were performed on physiological and environmental data, and Lake Louise Acute Mountain Sickness Scores (LLS) to establish an algorithm for AMS risk analysis. The individual R2 coefficients of determination between the LLS and the measured altitude, ambient temperature, atmospheric pressure, relative humidity, climbing speed, heart rate, blood oxygen saturation (SpO2), heart rate variability (HRV), were 0.1, 0.23, 0, 0.24, 0, 0.24, 0.27, and 0.35 respectively; incorporating all aforementioned variables, the R2 coefficient is 0.62. The bagged trees classifier achieved favorable classification results, yielding a model sensitivity, specificity, accuracy, and area under receiver operating characteristic curve of 0.999, 0.994, 0.998, and 1, respectively. CONCLUSION: The experiment results indicate the use of machine learning multivariate analysis have higher AMS prediction accuracies than analyses utilizing single varieties. The developed AMS evaluation model can serve as a reference for the future development of wearable devices capable of providing timely warnings of AMS risks to hikers.

General deep learning model for detecting diabetic retinopathy.

BACKGROUND: Doctors can detect symptoms of diabetic retinopathy (DR) early by using retinal ophthalmoscopy, and they can improve diagnostic efficiency with the assistance of deep learning to select treatments and support personnel workflow. Conventionally, most deep learning methods for DR diagnosis categorize retinal ophthalmoscopy images into training and validation data sets according to the 80/20 rule, and they use the synthetic minority oversampling technique (SMOTE) in data processing (e.g., rotating, scaling, and translating training images) to increase the number of training samples. Oversampling training may lead to overfitting of the training model. Therefore, untrained or unverified images can yield erroneous predictions. Although the accuracy of prediction results is 90%-99%, this overfitting of training data may distort training module variables. RESULTS: This study uses a 2-stage training method to solve the overfitting problem. In the training phase, to build the model, the Learning module 1 used to identify the DR and no-DR. The Learning module 2 on SMOTE synthetic datasets to identify the mild-NPDR, moderate NPDR, severe NPDR and proliferative DR classification. These two modules also used early stopping and data dividing methods to reduce overfitting by oversampling. In the test phase, we use the DIARETDB0, DIARETDB1, eOphtha, MESSIDOR, and DRIVE datasets to evaluate the performance of the training network. The prediction accuracy achieved to 85.38%, 84.27%, 85.75%, 86.73%, and 92.5%. CONCLUSIONS: Based on the experiment, a general deep learning model for detecting DR was developed, and it could be used with all DR databases. We provided a simple method of addressing the imbalance of DR databases, and this method can be used with other medical images.

Lemnalol Modulates the Electrophysiological Characteristics and Calcium Homeostasis of Atrial Myocytes.

Sepsis, an inflammatory response to infection provoked by lipopolysaccharide (LPS), is associated with high mortality, as well as ischemic stroke and new-onset atrial arrhythmia. Severe bacterial infections causing sepsis always result in profound physiological changes, including fever, hypotension, arrhythmia, necrosis of tissue, systemic multi-organ dysfunction and finally death. LPS challenge-induced inflammatory responses during sepsis may increase the likelihood of the arrhythmogenesis. Lemnalol is known to possess potent anti-inflammatory effects. This study examined whether Lemnalol (0.1 µM) could modulate the electrophysiological characteristics and calcium homeostasis of atrial myocytes under the influence of LPS (1µg/mL). Under challenge with LPS, Lemnalol-treated LA myocytes, had a longer AP duration at 20%, 50% and 90% repolarization of the amplitude, compared to the LPS-treated cells. LPS-challenged LA myocytes showed increased late sodium current, Na+-Ca2+ exchanger current, transient outward current, rapid component of delayed rectifier potassium current, tumor necrosis factor-α, NF-κB and increased phosphorylation of ryanodine receptor (RyR), but a lower L-type Ca2+ current than the control LA myocytes. Exposure to Lemnalol reversed the LPS-induced effects. The LPS-treated and control groups of LA myocytes, with or without the existence of Lemnalol. showed no apparent alterations in the sodium current amplitude or Cav1.2 expression. The expression of sarcoendoplasmic reticulum calcium transport ATPase (SERCA2) was reduced by LPS treatment, while Lemnalol ameliorated the LPS-induced alterations. The phosphorylation of RyR was enhanced by LPS treatment, while Lemnalol attenuated the LPS-induced alterations. In conclusion, Lemnalol modulates LPS-induced alterations of LA calcium homeostasis and blocks the NF-κB pathways, which may contribute to the attenuation of LPS-induced arrhythmogenesis.

Excavatolide B Modulates the Electrophysiological Characteristics and Calcium Homeostasis of Atrial Myocytes.

Severe bacterial infections caused by sepsis always result in profound physiological changes, including fever, hypotension, arrhythmia, necrosis of tissue, systemic multi-organ dysfunction, and finally death. The lipopolysaccharide (LPS) provokes an inflammatory response under sepsis, which may increase propensity to arrhythmogenesis. Excavatolide B (EXCB) possesses potent anti-inflammatory effects. However, it is not clear whether EXCB could modulate the electrophysiological characteristics and calcium homeostasis of atrial myocytes. This study investigated the effects of EXCB on the atrial myocytes exposed to lipopolysaccharide. A whole-cell patch clamp and indo-1 fluorimetric ratio technique was employed to record the action potential (AP), ionic currents, and intracellular calcium ([Ca2+]i) in single, isolated rabbit left atrial (LA) cardiomyocytes, with and without LPS (1 µg/mL) and LPS + EXCB administration (10 µM) for 6 ± 1 h, in order to investigate the role of EXCB on atrial electrophysiology. In the presence of LPS, EXCB-treated LA myocytes (n = 13) had a longer AP duration at 20% (29 ± 2 vs. 20 ± 2 ms, p < 0.05), 50% (52 ± 4 vs. 40 ± 3 ms, p < 0.05), and 90% (85 ± 5 vs. 68 ± 3 ms, p < 0.05), compared to the LPS-treated cells (n = 12). LPS-treated LA myocytes showed a higher late sodium current, Na⁺/Ca2+ exchanger current, transient outward current, and delayed rectifier potassium current, but a lower l-type Ca2+ current, than the control LA myocytes. Treatment with EXCB reversed the LPS-induced alterations of the ionic currents. LPS-treated, EXCB-treated, and control LA myocytes exhibited similar Na⁺ currents. In addition, the LPS-treated LA myocytes exhibited a lower [Ca2+]i content and higher sarcoplasmic reticulum calcium content, than the controls. EXCB reversed the LPS-induced calcium alterations. In conclusion, EXCB modulates LPS-induced LA electrophysiological characteristics and calcium homeostasis, which may contribute to attenuating LPS-induced arrhythmogenesis.

Ma's bamboo-based medicinal moxibustion therapy of low back pain in lumbar disc herniation: study protocol for a randomized controlled trial.

BACKGROUND: Lumbar disc herniation (LDH) is a common and frequently occurring disease in clinics. Low back pain and sciatica are the presenting symptoms of LDH. To some extent, it can be considered that measures with the capability to improve low back pain or sciatica have the potential to treat LDH. Ma's bamboo-based medicinal moxibustion therapy can effectively reduce the degree of low back pain and has been widely used. Studies of small sample size have seen significant improvement on pain relief. The aim of this trial is to evaluate the clinical efficacy and safety of Ma's bamboo-based medicinal moxibustion therapy in the treatment of LDH low back pain. METHODS/DESIGN: The trial is a multicenter, randomized, parallel-group, non-inferiority study. Three hundred and twelve patients will be randomly assigned to a Ma's bamboo-based medicinal moxibustion group (n=156) and an acupuncture group (n=156). Patients in each group will receive treatment every day, 6 times a week, 12 times in total. Follow-up will be conducted 14 days after treatment. The primary outcome will be the visual analog scale(VAS) at baseline, after 6 times of treatment, end of treatment, and follow-up. The secondary outcomes will include Oswestry disability indexes (ODI), modified Japanese Orthopaedic Association low back pain (M-JOA) score, serum ß-endorphin (ß-EP), and serum substance P (SP). ß-EP and SP, as well as safety evaluation indexes (routine blood, liver, and kidney function and electrocardiogram), will be measure at baseline and after the end of treatment. The number, nature, and severity of adverse events will be recorded. DISCUSSION: The results of the trial will compare the efficacy of low back pain in LDH between Ma's bamboo-based medicinal moxibustion group and the acupuncture group and will be expected to make a systematic and objective evaluation of the clinical efficacy and safety of Ma's bamboo-based medicinal moxibustion therapy. TRIAL REGISTRATION: ChiCTR2000038725 . Registered on 29 September 2020.

[Exploration on effect mechanism of Miao medicinal acupuncture therapy in treatment of knee osteoarthritis based on TRPV ion channel].

OBJECTIVE: To observe the effect of Miao medicinal acupuncture therapy on transient receptor potential vanilloid (TRPV) channel in knee joint synovial tissue of the rabbits with knee osteoarthritis (KOA) model and to explore the mechanism of Miao medicinal acupuncture therapy in treatment of KOA. METHODS: Of 34 New Zealand male rabbits, 6 rabbits were selected randomly as the normal group. KOA model was established in the rest rabbits by injecting a mixture of papain and L-cysteine in right knee joints. The 24 successfully modeled rabbits were randomized into a model group, a Miao medicinal acupuncture therapy group, a dermal needle group and a smearing group, 6 rabbits in each one. In the Miao medicinal acupuncture therapy group, Miao medicinal acupuncture therapy was adopted, in which, the roller type of dermal needle was used on the surface of right knee joint [a rectangle shape formed by "Xuehai" (SP 10), "Liangqiu" (ST 34), "Yanglingquan" (GB 34) and "Yinlingquan" (SP 9)], rolling in a "" shape, on which, Miao medicinal solution was smeared in advance. In the dermal needle group, the rolling stimulation was exerted on the right the right knee joint surface with the roller type of dermal needle. In the smearing group, Miao medicinal solution was smeared on the right knee joint surface. The intervention was given once every two days, 3 times weekly and the intervention was exerted consecutively for 4 weeks. Successively, on day 1, 21, 28, 35, 42 and 49 of experiment, paw withdrawal threshold (von Frey threshold) after mechanical stimulation was detected in the rabbits. HE staining was adopted to observe the histomorphological changes of the right knee joint cartilage in the rabbits. ELISA was used to determine the contents of interleukin-1 (IL-1ß) and tumor necrosis factor-α (TNF-α) in the right knee synovial fluid. Western blot method and real-time PCR were used to determine the relative expressions of protein and mRNA of TRPV1 and TRPV4 in knee synovial tissue of the rabbits. RESULTS: Compared with the normal group, on day 49 of experiment, von Frey threshold was reduced significantly in the rabbits of the model group (P<0.01), the integrety of cartilage surface of knee joint was seriously damaged, the structural layers were disordered, the chondrocytes were clustered, the tide lines were distorted and the matrix staining disappeared. The contents of IL-1ß and TNF-α of the right knee synovial fluid were increased significantly (P<0.01), and the relative expressions of protein and mRNA of TRPV1 and TRPV4 in the synovial tissue of the right knee were increased significantly (P<0.01). Compared with the model group, on day 49 of experiment, von Frey threshold was increased significantly in each of the Miao medicinal acupuncture therapy group, the dermal needle group and the smearing group (P<0.01). The right knee joint cartilage was complete in morphology, the structure clear was in layer, the cells were arranged in order and the matrix staining was uniform. The contents of IL-1ß and TNF-α of the right knee synovial fluid were reduced significantly (P<0.01, P<0.05), and the relative expressions of protein and mRNA of TRPV1 and TRPV4 in the synovial tissue of the right knee were reduced significantly (P<0.01). Compared with either the dermal needle group or the smearing group, on day 49 of experiment, von Frey threshold was increased significantly in the Miao medicinal acupuncture therapy group (P<0.01). The right knee joint cartilage surface was even, the structure layers were clear. The contents of IL-1ß and TNF-α of the right knee synovial fluid were reduced significantly (P<0.05), and the relative expressions of protein of TRPV1 and TRPV4 in the synovial tissue were reduced (P<0.01, P<0.05). Compared with the smearing group, the relative expression of TRPV1 mRNA in the synovial tissue was reduced significantly in the Miao medicinal acupuncture therapy group (P<0.05). CONCLUSION: Miao medicinal acupuncture therapy plays a role in treatment of KOA probably through inhibiting the expressions of IL-1ß and TNF-α of knee synovial fluid and down-regulating the expressions of protein and mRNA of TRPV1 and TRPV4 in knee synovial tissue.

Wireless Stimulus-on-Device Design for Novel P300 Hybrid Brain-Computer Interface Applications.

Improving the independent living ability of people who have suffered spinal cord injuries (SCIs) is essential for their quality of life. Brain-computer interfaces (BCIs) provide promising solutions for people with high-level SCIs. This paper proposes a novel and practical P300-based hybrid stimulus-on-device (SoD) BCI architecture for wireless networking applications. Instead of a stimulus-on-panel architecture (SoP), the proposed SoD architecture provides an intuitive control scheme. However, because P300 recognitions rely on the synchronization between stimuli and response potentials, the variation of latency between target stimuli and elicited P300 is a concern when applying a P300-based BCI to wireless applications. In addition, the subject-dependent variation of elicited P300 affects the performance of the BCI. Thus, an adaptive model that determines an appropriate interval for P300 feature extraction was proposed in this paper. Hence, this paper employed the artificial bee colony- (ABC-) based interval type-2 fuzzy logic system (IT2FLS) to deal with the variation of latency between target stimuli and elicited P300 so that the proposed P300-based SoD approach would be feasible. Furthermore, the target and nontarget stimuli were identified in terms of a support vector machine (SVM) classifier. Experimental results showed that, from five subjects, the performance of classification and information transfer rate were improved after calibrations (86.00% and 24.2 bits/ min before calibrations; 90.25% and 27.9 bits/ min after calibrations).

Systematic node management mechanism using ZigBee-based real-time vital sign information monitoring system to manage large numbers of patients.

BACKGROUND: Local hospitals must deal with large numbers of patients during mass casualty incidents, and the wireless sensor networks (WSNs) can help in these situations by monitoring vital signs. Conventional ZigBee nodes can obtain the ID of a device by assigning a unique 16-bit short address or by burning firmware into an IC. These methods tend to complicate node management and lack portability. OBJECTIVE: The study developed a node management mechanism to deal with a large number of patients in real-time, through the wireless monitoring of physiological signals. The mechanism proposed for the ZigBee WSN is based on a three-layer (Coordinator, Control Router, and End Device) tree topology. METHODS: The proposed system includes a node deployment process to formulate a ZigBee WSN as a tree topology, an algorithm to automatically number ZigBee nodes for monitoring and control system (MCS), and an algorithm to automatically obtain the short addresses of nodes for data collection. Specifically, an algorithm automatically collects data from ZigBee nodes for display on a computer graphical user interface (GUI). We also developed a reliable data transmission method capable of resolving the problem of packet loss. RESULTS: The proposed method has been applied in a local hospital. Our research findings provide a valuable reference for the development of ZigBee-based MCS. CONCLUSIONS: The proposed node management mechanism is faster, more reliable, and more intuitive to use, than traditional methods.

Epigallocatechin-3-gallate modulates arrhythmogenic activity and calcium homeostasis of left atrium.

BACKGROUND: Atrial fibrillation (AF) is the commonest sustained arrhythmia, and increases the risk of stroke, heart failure, and mortality. Calcium (Ca2+) overload and oxidative stress are thought to participate in the pathogenesis of AF. Epigallocatechin-3-gallate (EGCG) has an antioxidative effect and been shown to be beneficial in promoting cardiovascular health. However, it is not clear if EGCG directly modulates the electrophysiological characteristics and Ca2+ homeostasis of the left atrium (LA). METHODS AND RESULTS: Conventional microelectrodes, whole-cell patch-clamp, and Fluo-3 fluorometric ratio technique were performed using the isolated rabbit LA preparations or isolated single LA cardiomyocytes before and after EGCG treatment. EGCG (0.01, 0.1, 1, and 10µM) which concentration-dependently decreased the APD20 by 13±8%, 25±5%, 31±6%, and 37±5%, APD50 by 9±8%, 22±6%, 32±7%, and 40±4%, and APD90 by 2±12%, 9±8%, 24±10%, and 34±5% in LA preparations. EGCG (0.1µM) decreased the late sodium (Na+) current, L-type Ca2+ current, nickel-sensitive Na+-Ca2+ exchanger current, and transient outward current, but did not change the Na+ current and ultra-rapid delayed rectifier potassium current in LA cardiomyocytes. EGCG decreased intracellular Ca2+ transient and sarcoplasmic reticulum Ca2+ content in LA cardiomyocytes. Furthermore, EGCG decreased isoproterenol (ISO, 1µM)-induced burst firing. KT5823 (1µM) or KN93 (1µM) decreased the incidences of ISO-induced LA burst firing, which became lower with EGCG treatment. H89 (10µM) and KN92 (1µM) did not suppress the incidence of ISO-induced LA burst firing. However, EGCG decreased the incidences of ISO-induced LA burst firing in the presence of H89 or KN92. CONCLUSION: EGCG directly regulates LA electrophysiological characteristics and Ca2+ homeostasis, and suppresses ISO-induced atrial arrhythmogenesis through inhibiting Ca2+/calmodulin or cGMP-dependent protein kinases.