Fluorescent Droplet Cytometry for On-Cell Phenotype Tracking.

Profiling the heterogeneous phenotypes of live cancer cells is a key capability that requires single-cell analysis. However, acquiring information at the single-cell level for live cancer cells is challenging when small collections of cells are being targeted. Here, we report single-cell analysis for low abundance cells enabled by fluorescent droplet cytometry (FDC), an approach that uses a biomarker-specific enzymatic fluorescent assay carried out using a droplet microfluidic platform. FDC utilizes DNA-functionalized antibodies in droplets to achieve specific on-cell target detection and enables characterization and profiling of live cancer cells with single-cell resolution based on their surface phenotype. Using this approach, we achieve live-cell phenotypic profiling of multiple surface markers acquired with small (<40 cells) collections of cells.

Visualizing mRNA in live mammalian cells.

mRNAs play essential role in gene expression and regulation. Direct visualization of specific intracellular mRNAs can provide key information about mRNA distribution and dynamics that accelerating study of their functionality. This highlights the necessity of characterizing mRNA in mammalian cells with good spatial and temporal resolution. Live cell imaging and tracking of mRNA remains a challenge due to their relatively short half-life and lack of versatile imaging probes. This review provides an overview of current methods of live-cell mRNA imaging, including exogenous hybridization probes and genetically encoded strategies. It then focuses on the development and application of aptamer initiated fluorescence complementation (AiFC) method for real-time imaging of mammalian mRNAs and includes a detailed protocol for observing ß-actin mRNA in HeLa cells.

A Real Time QRS Detection Algorithm Based on ET and PD Controlled Threshold Strategy.

As one of the important components of electrocardiogram (ECG) signals, QRS signal represents the basic characteristics of ECG signals. The detection of QRS waves is also an essential step for ECG signal analysis. In order to further meet the clinical needs for the accuracy and real-time detection of QRS waves, a simple, fast, reliable, and hardware-friendly algorithm for real-time QRS detection is proposed. The exponential transform (ET) and proportional-derivative (PD) control-based adaptive threshold are designed to detect QRS-complex. The proposed ET can effectively narrow the magnitude difference of QRS peaks, and the PD control-based method can adaptively adjust the current threshold for QRS detection according to thresholds of previous two windows and predefined minimal threshold. The ECG signals from MIT-BIH databases are used to evaluate the performance of the proposed algorithm. The overall sensitivity, positive predictivity, and accuracy for QRS detection are 99.90%, 99.92%, and 99.82%, respectively. It is also implemented on Altera Cyclone V 5CSEMA5F31C6 Field Programmable Gate Array (FPGA). The time consumed for a 30-min ECG record is approximately 1.3 s. It indicates that the proposed algorithm can be used for wearable heart rate monitoring and automatic ECG analysis.

Nanoparticle-Mediated Capture and Electrochemical Detection of Methicillin-Resistant Staphylococcus aureus.

The spread of antibiotic-resistant bacteria poses a global threat to public health. Conventional bacterial detection and identification methods often require pre-enrichment and/or sample preprocessing and purification steps that can prolong diagnosis by days. Methicillin-resistant Staphylococcus aureus (MRSA) is one of the most widespread antibiotic-resistant bacteria and is the leading cause of hospital-acquired infections. Here, we have developed a method to specifically capture and detect MRSA directly from patient nasal swabs with no prior culture and minimal processing steps using a microfluidic device and antibody-functionalized magnetic nanoparticles. Bacteria are captured based on antibody recognition of a membrane-bound protein marker that confers ß-lactam antibiotic resistance. MRSA identification is then achieved by the use of a strain-specific antibody functionalized with alkaline phosphatase for electrochemical detection. This approach ensures that only those bacteria of the target strain and resistance profile are measured. The method has a limit of detection of 845 CFU/mL and excellent discrimination against high concentrations of common nontarget nasal flora with a turnaround time of under 4.5 h. This detection method was successfully validated using clinical nasal swab specimens ( n = 30) and has the potential to be tailored to various bacterial targets.

Large volume of water samples introduced in dispersive liquid-liquid microextraction for the determination of 15 triazole fungicides by gas chromatography-tandem mass spectrometry.

A novel method of large volume of water samples directly introduced in dispersive liquid-liquid microextraction was developed, which is based on ultrasound/manual shaking-synergy-assisted emulsification and self-generating carbon dioxide gas (CO2) breaking down the emulsion for the determination of 15 triazole fungicides by gas chromatography-tandem mass spectrometry. This technique makes low-density extraction solvent toluene (180 µL) dissolve in 200 mL of samples containing 0.05 mol L(-1) of HCl and 5 % of NaCl (w/v) to form a well emulsion by synergy of ultrasound and manual shaking, and injects NaHCO3 solution (1.0 mol L(-1)) to generate CO2 achieving phase separation with the assistance of ultrasound. The entire process is accomplished within 8 min. The injection of NaHCO3 to generate CO2 achieves phase separation that breaks through the centrifugation limited large volume aqueous samples. In addition, the device could be easily cleaned, and this kind of vessel could be reconfigured for any volume of samples. Under optimal conditions, the low limits of detection ranging from 0.7 to 51.7 ng L(-1), wide linearity, and enrichment factors obtained were in the range 924-3669 for different triazole fungicides. Southern end of the Beijing-Hangzhou Grand Canal water (Hangzhou, China) was used to verify the applicability of the developed method. Graphical Abstract Flow chart of ultrasound/manual shaking-synergy-assisted emulsification and self-generating carbon dioxide gas breaking down the emulsion.

[Study on Hydrothermal Preparation and Luminescence Properties of Luminescent Material BaSrMg(PO4)2:Eu³âº].

Eu³âº doped BaSrMg (PO4)2 were prepared by a hydrothermal method. The crystal structure and morphology of BaSrMg(PO4)2:Eu³âº phosphor were characterized by X-ray powder diffraction (XRD) and field emission scanning electron microscopy (FESEM). The effects of different pH values (5, 6, 7 and 8) and different reaction temperatures (120, 140, 160, 180 and 200 °C) on the crystal structure and morphology of BaSrMg(PO4)2:Eu³âº phosphor were studied in this paper. The results of XRD indicate that diffraction peaks are sharp and strong only when pH value is 6, meanwhile the FESEM shows the morphology is regular-shaped. The XRD patterns show amorphous halos superimposed with several weak sharp peaks for the samples preparing under the pH values of 5, 7 and 8. It indicates that these three samples are solid solution or mixed phases, which are in accord with the results of FESEM. From the fluorescence spectra, the peaks in the excitation spectra were assigned to the transition from 7F0 to 5D4, 5L8, 5L6 and 5D2, while the peaks of emission spectra corresponding to the transition of 5D1 --> 7F1 and 5D0-->7Fj (J = 0, 1, 2, 3 and 4). The strongest emission peak of the optimized phosphor located at 613 nm (5D0--> 7F2), excited by the main excitation peak with wavelength of 394 nm. The splitting of the emission peaks changes depends on pH values and temperatures, which indicating that luminescence properties is closely related to the crystal structure and morphology of particles.

[Detection of Ethoprophos Using SERS Coupled with Magnetic Fe3O4/Ag Composite Materials].

The magnetic Fe3O4/Ag composite materials were synthesized by reducing AgNO3 with sodium citrate in the presence of Fe3O4 which were prepared by co-precipitation firstly. The enrichment and extraction of ethoprophos assembled on Fe3O4/Ag were achieved with the applied magnetic field. The different concentrations of ethoprophos adsorbed on Fe3O4/Ag were analyzed by SERS and it was showed that the trace analysis of ethoprophos had been established, while the enhancement factor of probe molecules on Fe3O4/Ag was 1. 48 X 10(5). The structure and morphology of Fe3O4/Ag were characterized by UV-Vis, EDX and TEM. Compared with Ag, the UV-Vis absorption peak of Fe3O4/Ag shifted from 417 to 369 nm, and the UV-Vis of Fe3O4 almost had no characteristic absorption peak in this region. At the same time, it was showed that the surface properties of Fe3O4/Ag changed with Raman enhancement effect during the aggregation process of Ag around the surface of Fe3O4. Further EDX images of micro area element analysis suggested that the chemical composition of products were Ag, Fe and O while the Cu peak was from the copper mesh. In addition, TEM images indicated that the average particle size of Fe3O4 was between 30 and 60 nm with shape tended to be spherical. And the silver nanoparticles were attached to the Fe3O4 particles and agglomeration occured. Density functional theory calculations which can be applied to qualitative judgment of molecule was carried out to obtain the molecular optimization structure and theoretical Raman spectra. It was found that the stabilized SERS signals were detected under the saturated adsorption equilibrium after 15 min. Finally, Raman response of ethoprophos was achieved with lower than 2 X 10(-8) mol . L-1 , indicatint that the established method had reached the requirements of ethoprophos residues detection and could be used for analysis of sulfur-containing organophosphorus pesticide.

[Chemical characteristics of Nickel(II) 2,3-dimercaptosuccinate complexes modified on gold electrodes].

2,3-Dimercaptosuccinic acid (DMSA) and its complex Nickel(II) 2,3-Dimercaptosuccinate (DMSA-Ni(II)) were assembled on the electrochemically activated Au electrode and the modified films were investigated by ex-situ SERS, SEM and electro-reduction. The coordination ratio (1:2) and stability constant (4.716 X 10(7)) of DMSA and Nickel(II) were firstly confirmed by UV. Then, the SERS spectra showed that DMSA was absorbed on the surface of activated Au and the coordinate bond of DMSA-Ni(II) was existed as S-Ni-O. Meanwhile, two reduction peaks of DMSA-Ni(II) on Au electrode were presented at negative potential. The first peak (from positive to negative values) was related to the RS-Au group which was produced by the mutual absorption of DMSA and Au. The second reduction peak was related to Nickel(II). Finally, it was further confirmed by calculating the surface coverage and comparison of molecular area. And it was found that the molecular arrangement of DM-SA-Ni(II) on Au electrode generated as monolayer which was dispersed or connected partly instead of concentrated or overlapped.

Protective effects of chlorogenic acid on the intestinal barrier of broiler chickens: an immunological stress model study.

This study was conducted to investigate the protective effects of chlorogenic acid (CGA) on inflammatory responses and intestinal health of lipopolysaccharide (LPS)-challenged broilers. One hundred and forty-four 1-day-old male broiler chicks were divided into 3 groups with 6 replicates of 8 birds each. The groups were as follows: 1) Control group: birds fed a basal diet; 2) LPS group: LPS-challenged birds fed a basal diet; 3) CGA group: LPS-challenged birds fed a CGA-supplemented diet. The LPS was intraperitoneally administered at a dose of 1 mg/kg of body weight. CGA increased the weight gain and feed intake of LPS-challenged birds by 37.05% and 24.29%, respectively (P < 0.05). CGA also alleviated LPS-induced inflammation, as evidenced by lower levels of pro-inflammatory cytokines in the serum and jejunum (tumor necrosis factor-α, interferon-γ, interleukin-1ß, and interleukin-6), and the decreased myeloperoxidase activity in the jejunum (P < 0.05). These effects were accompanied by a decrease in the mRNA abundance of toll-like receptor 4 and myeloid differentiation factor 88 and an inhibition of nuclear factor kappa-B translocation in the jejunum (P < 0.05). CGA reduced circulating diamine oxidase activity and levels of D-lactate and endotoxin, and positively regulated the expression of jejunal claudin-3 and zonula occludens-1 in LPS-challenged broilers (P < 0.05). Compared to the LPS group, CGA reduced the apoptotic rate of epithelial cells and cytochrome c concentration in the jejunum, and normalized the expression of genes responsible for proliferation and apoptosis in jejunal epithelial cells, including cysteine aspartate-specific protease-9, B cell lymphoma-2, and proliferating cell nuclear antigen (P < 0.05). Furthermore, CGA normalized the altered phosphorylation of protein kinase B and glycogen synthase kinase-3ß, as well as the translocation of nuclear ß-catenin in the jejunum of LPS-challenged broilers (P < 0.05). These results suggested that CGA supplementation improved growth performance, alleviated inflammation, and helped maintain intestinal integrity and barrier function in LPS-challenged broilers, possibly through the regulation of the toll-like receptor 4/nuclear factor kappa-B and protein kinase B/Wnt/ß-catenin pathways.

Risk of Esophageal Adenocarcinoma After Bariatric Surgery: A Meta-Analysis of Retrospective Studies.

PURPOSE: This study aims to systematically review and meta-analyze the evidence on the risk of esophageal adenocarcinoma (EAC) following metabolic and bariatric surgery (MBS). MATERIALS AND METHODS: A systematic literature search was conducted on the China National Knowledge Infrastructure (CNKI), Wanfang, EMBASE, MEDLINE, Web of Science, The Cochrane Library, and PubMed databases. Meta-analysis utilized odds ratios (ORs) and 95% confidence intervals (CIs) to analyze the correlation between MBS and the risk of EAC. Meta-analysis was performed using STATA software (version 12.0). RESULTS: Fourteen studies involving patients with obesity undergoing bariatric surgery and control groups receiving conventional treatment were included. The meta-analysis indicated a reduction in the overall incidence of esophageal cancer after bariatric surgery (OR = 0.69, 95% CI: 0.51-0.95, P = 0.022). Subgroup analysis results demonstrated a decreased risk of EAC in European patients with obesity undergoing MBS treatment (OR: 0.60, 95% CI: 0.38-0.95, P = 0.028). In studies with a sample size greater than or equal to 100,000 patients, the risk of EAC in patients with obesity undergoing MBS was significantly lower than the non-surgery group (OR: 0.59, 95% CI: 0.42-0.83, P = 0.003). Articles published before 2020 and those published in 2020 or earlier showed a significant difference in the incidence of EAC between the surgery and non-surgery groups (OR: 0.57, 95% CI: 0.43-0.75, P < 0.001). The risk of EAC in patients with obesity with a follow-up time of less than 5 years was statistically significant (OR: 0.46, 95% CI: 0.25-0.82, P = 0.009). CONCLUSION: Our meta-analysis results suggest a reduced risk of esophageal cancer in patients with obesity after bariatric surgery. PROSPERO REGISTRATION: CRD 42024505177.

Research advances in serum chitinase-3-like protein 1 in liver fibrosis.

While liver fibrosis remains a serious, progressive, chronic liver disease, and factors causing damage persist, liver fibrosis may develop into cirrhosis and liver cancer. However, short-term liver fibrosis is reversible. Therefore, an early diagnosis of liver fibrosis in the reversible transition phase is important for effective treatment of liver diseases. Chitinase-3-like protein 1 (CHI3L1), an inflammatory response factor that participates in various biological processes and is abundant in liver tissue, holds promise as a potential biomarker for liver diseases. Here, we aimed to review research developments regarding serum CHI3L1 in relation to the pathophysiology and diagnosis of liver fibrosis of various etiologies, providing a reference for the diagnosis, treatment, and prognosis of liver diseases.

Sleep and mental health during the COVID-19 pandemic: findings from an online questionnaire survey in China.

Introduction: The online study investigated the sleep, psychological conditions, and risk factors during the wave of transmission of COVID-19 since December 7, 2022. Methods: We distributed questionnaires through networking mediums to residents to gather information about COVID-19 infection, sleep, and mental status. Results: During the extraordinary period in China, 91.9% of 1094 participants claimed to be infected with COVID-19, 36.8% reported poor sleep quality, 75.9% reported anxiety, and 65.5% reported depression. In retrospect, people have experienced lower sleep quality, longer sleep latency, enhanced rising time, and decreased sleep efficiency after the infection wave. After adjusting confounding factors, the elderly, women, urban residents, people with comorbidity, anxiety, depression, stress state, and COVID-19 infection have high risks for sleep disorders during the period. Discussion: The survey indicates that sleep disturbance caused by COVID-19 involves multiple dimensions, such as physiology, psychology, and society. The COVID-19 infection-related sleep problem should be taken seriously. Apart from conventional treatment, psychological issues of insomnia can not be ignored.

CPSS: Fusing consistency regularization and pseudo-labeling techniques for semi-supervised deep cardiovascular disease detection using all unlabeled electrocardiograms.

BACKGROUND AND OBJECTIVE: Deep learning usually achieves good performance in the supervised way, which requires a large amount of labeled data. However, manual labeling of electrocardiograms (ECGs) is laborious that requires much medical knowledge. Semi-supervised learning (SSL) provides an effective way of leveraging unlabeled data to improve model performance, providing insight for solving this problem. The objective of this study is to improve the performance of cardiovascular disease (CVD) detection by fully utilizing unlabeled ECG. METHODS: A novel SSL algorithm fusing consistency regularization and pseudo-labeling techniques (CPSS) is proposed. CPSS consists of supervised learning and unsupervised learning. For supervised learning, the labeled ECGs are mapped into prediction vectors by the classifier. The cross-entropy loss function is used to optimize the classifier. For unsupervised learning, the unlabeled ECGs are weakly and strongly augmented, and a consistency loss is used to minimize the difference between the classifier's predictions for the two augmentations. Pseudo-labeling techniques include positive pseudo-labeling (PL) and ranking-based negative pseudo-labeling (RNL). PL introduces pseudo-labels for data with high prediction confidence. RNL assigns negative pseudo-labels to the lower-ranked categories in the prediction vectors to leverage data with low prediction confidence. In this study, VGGNet and ResNet are used as classifiers, which are jointly optimized by labeled and unlabeled ECGs. RESULTS: CPSS has been validated on several databases. With the same number of labeled ECGs (10%), it improves the accuracies over pure supervised learning by 13.59%, 4.60%, and 5.38% in the CPSC2018, PTB-XL, and Chapman databases, respectively. CPSS achieves comparable results to the fully supervised method with only 10% of labeled ECGs, which reduces the labeling workload by 90%. In addition, to verify the practicality of CPSS, a cardiovascular disease monitoring system is designed by heterogeneously deploying the trained classifiers on an SoC (system-on-a-chip), which can detect CVD in real time. CONCLUSION: The results of this study indicate that the proposed CPSS can significantly improve the performance of CVD detection using unlabeled ECG, which reduces the burden of ECG labeling in deep learning. In addition, the designed monitoring system makes the proposed CPSS promising for real-world applications.

Short-Chain Sulfur Confined into Nitrogen-Doped Hollow Carbon Nanospheres for High-Capacity Potassium Storage.

Carbon-based materials are one of the ideal negative electrode materials for potassium ion batteries. However, the limited active sites and sluggish diffusion ion kinetics still hinder its commercialization process. To address these problems, we design a novel carbon composite anode, by confining highly reactive short-chain sulfur molecules into nitrogen-doped hollow carbon nanospheres (termed SHC-450). The formation process involves the controlled synthesis of hollow polyaniline (PANI) nanospheres as precursors via an Ostwald ripening mechanism and subsequent sulfuration treatment. The high content of constrained short-chain sulfur molecules (20.94 wt%) and considerable N (7.15 wt%) ensure sufficient active sites for K+ storage in SHC-450. Accordingly, the SHC-450 electrode exhibits a high reversible capacity of 472.05 mAh g-1 at 0.1 A g-1 and good rate capability (172 mAh g-1 at 2 A g-1). Thermogravimetric analysis shows that SHC-450 has impressive thermal stability to withstand a high temperature of up to 640 °C. Ex situ spectroscopic characterizations reveal that the short-chain sulfur provides high capacity through reversible formation of K2S. Moreover, its special hollow structure not only provides ample space for highly active short-chain sulfur reactants but also effectively mitigates volume expansion during the sulfur conversion process. This work offers new perspectives on enhanced K+ storage performance from an interesting anode design and the space-limited domain principle.

TLR3 activation enhances abscopal effect of radiotherapy in HCC by promoting tumor ferroptosis.

Radiotherapy (RT) has been reported to induce abscopal effect in advanced hepatocellular carcinoma (HCC), but such phenomenon was only observed in sporadic cases. Here, we demonstrated that subcutaneous administration of Toll-like receptor 3 (TLR3) agonist poly(I:C) could strengthen the abscopal effect during RT through activating tumor cell ferroptosis signals in bilateral HCC subcutaneous tumor mouse models, which could be significantly abolished by TLR3 knock-out or ferroptosis inhibitor ferrostatin-1. Moreover, poly(I:C) could promote the presentation of tumor neoantigens by dendritic cells to enhance the recruitment of activated CD8+ T cells into distant tumor tissues for inducing tumor cell ferroptosis during RT treatment. Finally, the safety and feasibility of combining poly(I:C) with RT for treating advanced HCC patients were further verified in a prospective clinical trial. Thus, enhancing TLR3 signaling activation during RT could provide a novel strategy for strengthening abscopal effect to improve the clinical benefits of advanced HCC patients.

[Chemical adsorption of thiosalicylic acid on silver surface investigated by surface-enhanced Raman scattering].

Surface-enhanced Raman scattering (SERS) of thiosalicylic acid (TSA) with sulfydryl group was investigated on the surface of electrochemically roughed silver electrode. The result shows that SERS enhancement effect was relative to the concentration and pH, and 1 x 10(-3) mol x L(-1) and pH 4 were the optimal condition. While the concentration increased, the enhancement effect decreased quickly because of steric hindrance. S--Ag peak position by the absorption of TSA was basically consistent, but pH significantly affected its intensity. The distribution and mechanism of TSA at different pH were further investigated. It was showed that TSA was adsorbed on the active silver surface via the sulfydryl group without H of neutral C4H4 (COOH)SH molecule. Competitive adsorption of negative valence C4H4 (COO-) SH and OH- may bring non-SERS under the strong base condition. At the same time, the sulfydryl group significantly influenced the carboxyl vibration peak's change and the distribution of electron cloud of benzene ring conjugate system.

Effects of dietary supplementation with chlorogenic acid on growth performance, antioxidant capacity, and hepatic inflammation in broiler chickens subjected to diquat-induced oxidative stress.

This study was conducted to investigate the protective effects of chlorogenic acid (CGA) on broilers subjected to (DQ)-induced oxidative stress. In experiment 1, one hundred and ninety-two male one-day-old Ross 308 broiler chicks were distributed into 4 groups and fed a basal diet supplemented with 0, 250, 500, or 1,000 mg/kg CGA for 21 d. In experiment 2, an equivalent number of male one-day-old chicks were allocated to 4 treatments for a 21-d trial: 1) Control group, normal birds fed a basal diet; 2) DQ group, DQ-challenged birds fed a basal diet; and 3) and 4) CGA-treated groups: DQ-challenged birds fed a basal diet supplemented with 500 or 1,000 mg/kg CGA. The intraperitoneal DQ challenge was performed at 20 d. In experiment 1, CGA administration linearly increased 21-d body weight, and weight gain and feed intake during 1 to 21 d (P < 0.05). CGA linearly and/or quadratically increased total antioxidant capacity, catalase, superoxide dismutase, and glutathione peroxidase activities, elevated glutathione level, and reduced malondialdehyde accumulation in serum, liver, and/or jejunum (P < 0.05). In experiment 2, compared with the control group, DQ challenge reduced body weight ratio (P < 0.05), which was reversed by CGA administration (P < 0.05). DQ challenge increased serum total protein level, aspartate aminotransferase activity, and total bilirubin concentration (P < 0.05), which were normalized when supplementing 500 mg/kg and/or 1,000 mg/kg CGA (P < 0.05). DQ administration elevated hepatic interleukin-1ß, tumor necrosis factor-α, and interleukin-6 levels (P < 0.05), and the values of interleukin-1ß were normalized to control values when supplementing CGA (P < 0.05). DQ injection decreased serum superoxide dismutase activity, hepatic catalase activity, and serum and hepatic glutathione level, but increased malondialdehyde concentration in serum and liver (P < 0.05), and the values of these parameters (except hepatic catalase activity) were reversed by 500 and/or 1,000 mg/kg CGA. The results suggested that CGA could improve growth performance, alleviate oxidative stress, and ameliorate hepatic inflammation in DQ-challenged broilers.

A joint cross-dimensional contrastive learning framework for 12-lead ECGs and its heterogeneous deployment on SoC.

The utilization of unlabeled electrocardiogram (ECG) data is always a critical topic in artificial intelligence healthcare, as the manual annotation for ECG data is a time-consuming task that requires much medical expertise. The recent development of self-supervised learning, especially contrastive learning, has provided helpful inspirations to solve this problem. In this paper, a joint cross-dimensional contrastive learning algorithm for unlabeled 12-lead ECGs is proposed. Unlike existing studies about ECG contrastive learning, our algorithm can simultaneously exploit unlabeled 1-dimensional ECG signals and 2-dimensional ECG images. A cross-dimensional contrastive learning method enhances the interaction between 1-dimensional and 2-dimensional ECG data, resulting in a more effective self-supervised feature learning. Combining this cross-dimensional contrastive learning, a 1-dimensional contrastive learning with ECG-specific transformations is employed to constitute a joint model. To pre-train this joint model, a new hybrid contrastive loss balances the 2 algorithms and uniformly describes the pre-training target. In the downstream classification task, the features learned by our algorithm shows impressive advantages. Compared with other representative methods, it achieves a at least 5.99% increase in accuracy. For real-world applications, an efficient heterogenous deployment on a "system-on-a-chip" (SoC) is designed. According to our experiments, the model can process 12-lead ECGs in real-time on the SoC. Furthermore, this heterogenous deployment can achieve a 14 × faster inference than the pure software deployment on the same SoC. In summary, our algorithm is a good choice for unlabeled 12-lead ECG utilization, the proposed heterogenous deployment makes it more practical in real-world applications.

ST-ReGE: A Novel Spatial-Temporal Residual Graph Convolutional Network for CVD.

Recently, deep learning (DL) has enabled rapid advancements in electrocardiogram (ECG)-based automatic cardiovascular disease (CVD) diagnosis. Multi-lead ECG signals have lead systems based on the potential differences between electrodes placed on the limbs and the chest. When applying DL models, ECG signals are usually treated as synchronized signals arranged in Euclidean space, which is the abstraction and generalization of real space. However, conventional DL models typically merely focus on temporal features when analyzing Euclidean data. These approaches ignore the spatial relationships of different leads, which are physiologically significant and useful for CVD diagnosis because different leads represent activities of specific heart regions. These relationships derived from spatial distributions of electrodes can be conveniently created in non-Euclidean data, making multi-lead ECGs better conform to their nature. Considering graph convolutional network (GCN) adept at analyzing non-Euclidean data, a novel spatial-temporal residual GCN for CVD diagnosis is proposed in this work. ECG signals are firstly divided into single-channel patches and transferred into nodes, which will be connected by spatial-temporal connections. The proposed model employs residual GCN blocks and feed-forward networks to alleviate over-smoothing and over-fitting. Moreover, residual connections and patch dividing enable the capture of global and detailed spatial-temporal features. Experimental results reveal that the proposed model achieves at least a 5.85% and 6.80% increase in F1 over other state-of-the-art algorithms with similar parameters and computations in both PTB-XL and Chapman databases. It indicates that the proposed model provides a promising avenue for intelligent diagnosis with limited computing resources.

A fully-mapped and energy-efficient FPGA accelerator for dual-function AI-based analysis of ECG.

In this paper, a fully-mapped field programmable gate array (FPGA) accelerator is proposed for artificial intelligence (AI)-based analysis of electrocardiogram (ECG). It consists of a fully-mapped 1-D convolutional neural network (CNN) and a fully-mapped heart rate estimator, which constitute a complementary dual-function analysis. The fully-mapped design projects each layer of the 1-D CNN to a hardware module on an Intel Cyclone V FPGA, and a virtual flatten layer is proposed to effectively bridge the feature extraction layers and fully-connected layer. Also, the fully-mapped design maximizes computational parallelism to accelerate CNN inference. For the fully-mapped heart rate estimator, it performs pipelined transformations, self-adaptive threshold calculation, and heartbeat count on the FPGA, without multiplexed usage of hardware resources. Furthermore, heart rate calculation is elaborately analyzed and optimized to remove division and acceleration, resulting in an efficient method suitable for hardware implementation. According to our experiments on 1-D CNN, the accelerator can achieve 43.08× and 8.38× speedup compared with the software implementations on ARM-Cortex A53 quad-core processor and Intel Core i7-8700 CPU, respectively. For the heart rate estimator, the hardware implementations are 25.48× and 1.55× faster than the software implementations on the two aforementioned platforms. Surprisingly, the accelerator achieves an energy efficiency of 63.48 GOPS/W, which obviously surpasses existing studies. Considering its power consumption is only 67.74 mW, it may be more suitable for resource-limited applications, such as wearable and portable devices for ECG monitoring.