Mg-based implants with a sandwiched composite coating simultaneously facilitate antibacterial and osteogenic properties.

Insufficient antibacterial effects and over-fast degradation are the main limitations of magnesium (Mg)-based orthopedic implants. In this study, a sandwiched composite coating containing a triclosan (TCS)-loaded poly(lactic acid) (PLA) layer inside and brushite (DCPD) layer outside was prepared on the surface of the Mg-Nd-Zn-Zr (denoted as JDBM) implant. In vitro degradation tests revealed a remarkable improvement in the corrosion resistance and moderate degradation rate. The drug release profile demonstrated a controllable and sustained TCS release for at least two weeks in vitro. The antibacterial rates of the implant were all over 99.8% for S. aureus, S. epidermidis, and E. coli, demonstrating superior antibacterial effects. Additionally, this coated JDBM implant exhibited no cytotoxicity but improved cell adhesion and proliferation, indicating excellent cytocompatibility. In vivo assays were conducted by implant-related femur osteomyelitis and osseointegration models in rats. Few bacteria were attached to the implant surface and the surrounding bone tissue. Furthermore, the coated JDBM implant exhibited more new bone formation than other groups due to the synergistic biological effects of released TCS and Mg2+, revealing excellent osteogenic ability. In summary, the JDBM implant with the sandwiched composite coating could significantly enhance the antibacterial activities and osteogenic properties simultaneously by the controllable release of TCS and Mg2+, presenting great potential for clinical transformation.

Enhanced Adsorption Stability and Biofunction Durability with Phosphonate-Grafted, PEGylated Copolymer on Hydroxyapatite Surface.

Nonfouling surfaces are crucial in applications such as biosensors, medical implants, marine coatings, and drug delivery vehicles. However, their long-term coating stability and robust surface binding strength in physiological media remain challenging. Herein, a phosphonate-grafted, PEGylated copolymer on the hydroxyapatite (HA) surface is proposed to significantly improve the adsorption stability and thus enhance the biofunction durability accordingly. The phosphoryl (-PO3) grafted branch is employed in the functional polymer to facilitate attaching to the HA substrate. In addition, the polymer integrates the nonfouling polymer brushes of poly(ethylene glycol) (PEG) with the cell-adhesive moiety of cyclic Arg-Gly-Asp-d-Phe-Cys peptides (cRGD). A systematic study on the as-synthesized PEGylated graft copolymer indicates a synergistic binding mechanism of the NH2 and PO3 groups to HA, achieving a high surface coverage with desirable adsorption stability. The cRGD/PEGylated copolymers of optimized grafting architecture are proven to effectively adsorb to HA surfaces as a self-assembled copolymer monolayer, showing stability with minimal desorption even in a complex, physiological medium and effectively preventing nonspecific protein adsorption as examined with X-ray photoelectron spectroscopy (XPS) and a quartz crystal microbalance with dissipation (QCM-D). Direct adhesion assays further confirm that the enhanced coating stability and biofunction durability of the phosphonate-grafted, cRGD-PEGylated copolymer can considerably promote osteoblast attachment on HA surfaces, meanwhile preventing microbial adhesion. This research has resulted in a solution of self-assembly polymer structure optimization that exhibits stable nonfouling characteristics.

The correlation of the intestinal with pharyngeal microbiota in early neonates.

Introduction: The gut-lung axis has long been recognized as an important mechanism affecting intestinal and lung immunity. Still, few studies have examined the correlation between the intestinal and pharyngeal microbiota in early neonates, especially when feeding patterns are one of the main drivers of microbiota development. Methods: To explore the composition and function of intestinal and pharyngeal microbiota and to analyze the effect of limited formula feeding on the initial microbiota colonization in early full-term neonates, we characterized the stool and oropharyngeal microbiota of 20 healthy full-term newborns sampled on days 0 and 5-7 after birth using 16S rRNA gene sequencing. Based on the sequencing results, a comparison was made of the compositions and functions of the intestinal and oropharyngeal microbiota for analysis. Results and discussion: At the phylum level, Firmicutes, Actinobacteria, Proteobacteria, and Bacteroidetes were the most abundant in both niches. At the genus level, the species of pioneer bacteria were rich in the intestine and oropharynx but low in abundance on day 0. On days 5-7, Bifidobacterium (25.40%) and Escherichia-Shigella (22.16%) were dominant in the intestine, while Streptococcus (38.40%) and Staphylococcus (23.13%) were dominant in the oropharynx. There were eight core bacteria genera in the intestine and oropharynx on days 5-7, which were Bifidobacterium, Escherichia-Shigella, Staphylococcus, Streptococcus, Bacteroides, Parabacteroides, Rothia, and Acinetobacter. As indicated by PICRUSt analysis, on days 5-7, the intestinal microbiota was more predictive than the oropharyngeal microbiota in transcription, metabolism, cell motility, cellular processes and signaling, and organismal system function in the KEGG pathway. Compared to exclusive breastfeeding, limited formula feeding (40-60%) had no significant effect on the neonatal intestinal and oropharyngeal microbiota composition during the initial colonization period. Our results suggest that the initial colonization of microbiota is closely related to the ecological niche environment in the intestine and oropharynx, with their core microbiota being closely correlated. We found that early limited formula feeding could not significantly affect the initial colonization of microbiota in the intestine and oropharynx.

[Intestinal and pharyngeal microbiota in early neonates: an analysis based on high-throughput sequencing]. / åŸºäºŽé«˜é€šé‡æµ‹åºåˆ†æžæ—©æœŸæ–°ç”Ÿå„¿è‚ é“å’Œå’½éƒ¨å¾®ç”Ÿç‰©èŒç¾¤.

OBJECTIVES: To investigate the distribution characteristics and correlation of intestinal and pharyngeal microbiota in early neonates. METHODS: Full-term healthy neonates who were born in Shanghai Pudong New Area Maternal and Child Health Hospital from September 2021 to January 2022 and were given mixed feeding were enrolled. The 16S rRNA sequencing technique was used to analyze the stool and pharyngeal swab samples collected on the day of birth and days 5-7 after birth, and the composition and function of intestinal and pharyngeal microbiota were analyzed and compared. RESULTS: The diversity analysis showed that the diversity of pharyngeal microbiota was higher than that of intestinal microbiota in early neonates, but the difference was not statistically significant (P>0.05). On the day of birth, the relative abundance of Proteobacteria in the intestine was significantly higher than that in the pharynx (P<0.05). On days 5-7 after birth, the relative abundance of Actinobacteria and Proteobacteria in the intestine was significantly higher than that in the pharynx (P<0.05), and the relative abundance of Firmicutes in the intestine was significantly lower than that in the pharynx (P<0.05). At the genus level, there was no significant difference in the composition of dominant bacteria between the intestine and the pharynx on the day of birth (P>0.05), while on days 5-7 after birth, there were significant differences in the symbiotic bacteria of Streptococcus, Staphylococcus, Rothia, Bifidobacterium, and Escherichia-Shigella between the intestine and the pharynx (P<0.05). The analysis based on the database of Clusters of Orthologous Groups of proteins showed that pharyngeal microbiota was more concentrated on chromatin structure and dynamics and cytoskeleton, while intestinal microbiota was more abundant in RNA processing and modification, energy production and conversion, amino acid transport and metabolism, carbohydrate transport and metabolism, coenzyme transport and metabolism, and others (P<0.05). The Kyoto Encyclopedia of Genes and Genomes analysis showed that compared with pharyngeal microbiota, intestinal microbiota was more predictive of cell motility, cellular processes and signal transduction, endocrine system, excretory system, immune system, metabolic diseases, nervous system, and transcription parameters (P<0.05). CONCLUSIONS: The composition and diversity of intestinal and pharyngeal microbiota of neonates are not significantly different at birth. The microbiota of these two ecological niches begin to differentiate and gradually exhibit distinct functions over time.

Design of an Efficient CNN-based Cough Detection System on Lightweight FPGA.

Precisely and automatically detecting the cough sound is of vital clinical importance. Nevertheless, due to privacy protection considerations, transmitting the raw audio data to the cloud is not permitted, and therefore there is a great demand for an efficient, accurate, and low-cost solution at the edge device. To address this challenge, we propose a semi-custom software-hardware co-design methodology to help build the cough detection system. Specifically, we first design a scalable and compact convolutional neural network (CNN) structure that generates many network instances. Second, we develop a dedicated hardware accelerator to perform the inference computation efficiently, and then we find the optimal network instance by applying network design space exploration. Finally, we compile the optimal network and let it run on the hardware accelerator. The experimental results demonstrate that our model achieves 88.8% classification accuracy, 91.2% sensitivity, 86.5% specificity, and 86.5% precision, while the computation complexity is only 1.09M multiply-accumulation (MAC). Additionally, when implemented on a lightweight field programmable gate array (FPGA), the complete cough detection system only occupies 7.9K lookup tables (LUTs), 12.9K flip-flops (FFs), and 41 digital signal processing (DSP) slices, providing 8.3 GOP/s actual inference throughput and total power dissipation of 0.93 W. This framework meets the needs of partial application and can be easily extended or integrated into other healthcare applications.

MBFQuant: A Multiplier-Bitwidth-Fixed, Mixed-Precision Quantization Method for Mobile CNN-Based Applications.

Deploying Convolutional Neural Network (CNN)-based applications to mobile platforms can be challenging due to the conflict between the restricted computing capacity of mobile devices and the heavy computational overhead of running a CNN. Network quantization is a promising way of alleviating this problem. However, network quantization can result in accuracy degradation and this is especially the case with the compact CNN architectures that are designed for mobile applications. This paper presents a novel and efficient mixed-precision quantization pipeline, called MBFQuant. It redefines the design space for mixed-precision quantization by keeping the bitwidth of the multiplier fixed, unlike other existing methods, because we have found that the quantized model can maintain almost the same running efficiency, so long as the sum of the quantization bitwidth of the weight and the input activation of a layer is a constant. To maximize the accuracy of a quantized CNN model, we have developed a Simulated Annealing (SA)-based optimizer that can automatically explore the design space, and rapidly find the optimal bitwidth assignment. Comprehensive evaluations applying ten CNN architectures to four datasets have served to demonstrate that MBFQuant can achieve improvements in accuracy of up to 19.34% for image classification and 1.12% for object detection, with respect to a corresponding uniform bitwidth quantized model.

Automatic cough detection from realistic audio recordings using C-BiLSTM with boundary regression.

Automatic cough detection in the patients' realistic audio recordings is of great significance to diagnose and monitor respiratory diseases, such as COVID-19. Many detection methods have been developed so far, but they are still unable to meet the practical requirements. In this paper, we present a deep convolutional bidirectional long short-term memory (C-BiLSTM) model with boundary regression for cough detection, where cough and non-cough parts need to be classified and located. We added convolutional layers before the LSTM to enhance the cough features and preserve the temporal information of the audio data. Considering the importance of the cough event integrity for subsequent analysis, the novel model includes an embedded boundary regression on the last feature map for both higher detection accuracy and more accurate boundaries. We delicately designed, collected and labelled a realistic audio dataset containing recordings of patients with respiratory diseases, named the Corp Dataset. 168 h of recordings with 9969 coughs from 42 different patients are included. The dataset is published online on the MARI Lab website (https://mari.tongji.edu.cn/info/1012/1030.htm). The results show that the system achieves a sensitivity of 84.13%, a specificity of 99.82% and an intersection-over-union (IoU) of 0.89, which is significantly superior to other related models. With the proposed method, all the criteria on cough detection significantly increased. The open source Corp Dataset provides useful material and a benchmark for researchers investigating cough detection. We propose the state-of-the-art system with boundary regression, laying the foundation for identifying cough sounds in real-world audio data.

Microarray and Bioinformatics Analysis of Circular RNA Differential Expression in Newborns With Acute Respiratory Distress Syndrome.

Previous studies pointed out that a variety of microRNAs (miRNAs) are involved in the pathogenesis of neonatal acute respiratory distress syndrome (NARDS) and play different roles in the pathological process. However, there have been few studies reporting the connection between circular RNA (circRNA) and NARDS, so the expression profile of circRNAs in newborns with acute respiratory distress syndrome remains largely unknown. In the present study, 10 samples obtained from remaining clinical blood samples of newborns hospitalized in a neonatal ward of the First Affiliated Hospital of Nanjing Medical University from January 2020 to October 2020 were divided into the "NARDS" group and "non-NARDS" group according to the Montelux standard and then were analyzed in microarray, and 10 other samples collected from the same place and from January 1, 2021 to August 31, 2021, were used to do RT-qPCR experiment. circRNA expression profiles, in which 741 circRNAs were downregulated and 588 were upregulated, were screened with circRNA high-throughput sequencing. Subsequently, Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analysis of parent genes of the differentially expressed circRNAs revealed that these circRNAs may be related to the process of protein synthesis and metabolism in NARDS. Moreover, five circRNAs-hsa_circ_0058495, hsa_circ_0000367, hsa_circ_0005389, hsa_circ_0059571, and hsa_circ_0006608-were selected randomly among the top 10 circRNAs of the downregulated or upregulated expression profiles. Then, bioinformatics tools were used to predict correlative miRNA and its target genes, which were also subjected to the same bioinformatics analysis for further study. The top 30 enriched KEGG pathway analyses of the 125 target genes suggested that these target genes are widely involved in the synthesis and secretion of endocrine hormones, and the top 30 enriched GO terms based on the 125 target genes are also focused on the protein and DNA processing. Thus, the present results show that circRNAs could promote the inflammation of NARDS which may provide a new therapeutic direction and it can be used as molecular markers for early diagnosis of NARDS, but further molecular biology verification is needed to define the specific role of differentially expressed circRNAs in NARDS.

Effects of composition and hierarchical structures of calcium phosphate coating on the corrosion resistance and osteoblast compatibility of Mg alloys.

Magnesium and its alloys have been recently used in biomedical applications such as orthopedic implants, whereas the weak corrosion resistance undermines their clinical efficacy. Herein, to address this critical challenge, the preparation of hierarchically structured hydroxyapatite-based coatings was proposed. Compact coatings were fabricated on a Mg alloy through a facile two-step method of chemical deposition of brushite precursor and subsequent hydrothermal conversion. A series of HA-based coatings were obtained with kinetic conversion process with formation mechanism revealed. The hydroxyapatite coating demonstrated the greatest corrosion resistance for Mg in electrochemical and long-term immersion tests, especially against pitting corrosion, attributable to its compact structure, alkaline degradation environment and self-induced growth capacity. The in vitro cytocompatibility and osteoinductivity were dictated. Additionally, anti-corrosion mechanisms were compared among different coating compositions and structures, along with their correlation with cellular response. Our study brings hints for a tailored surface design for resorbable biomedical device applications.

Incorporation of Cerium Oxide into Hydroxyapatite Coating Protects Bone Marrow Stromal Cells Against H2O2-Induced Inhibition of Osteogenic Differentiation.

Oxidative stress exerts a key influence in osteoporosis in part by inhibiting osteogenic differentiation of bone marrow stromal cells (BMSCs). With their unique antioxidant properties and reported biocompatibility, cerium oxide (CeO2) ceramics exhibit promising potential for the treatment of osteoporosis resulting from oxidative stress. In this study, protective effects of CeO2-incorporated hydroxyapatite coatings (HA-10Ce and HA-30Ce) on the viability and osteogenic differentiation of H2O2-treated BMSCs were examined. CeO2-incorporated HA coatings enhanced cell viability and attenuated cell apoptosis caused by H2O2. An increase in CeO2 content in HA coatings better alleviated H2O2-induced inhibition of osteogenic differentiation by increasing alkaline phosphatase (ALP) activity, calcium deposition activity, and mRNA expression levels of osteogenesis markers runt-related transcription factor 2 (Runx2), ALP, and osteocalcin (OCN) in BMSCs. Furthermore, the H2O2-induced decrease of gene and protein expressions of ß-catenin and cyclin D1 in the Wnt/ß-catenin signaling pathway was successfully rescued by the CeO2 incorporated HA coatings. Besides, the decreased expression of receptor activator of nuclear factor kappa-B ligand (RANKL) and the increased ratio of osteoprotegerin (OPG)/RANKL in BMSCs on the CeO2-modified coatings was observed, indicating the inhibition of osteoclastogenesis. The above results were mediated by the antioxidant properties of CeO2. The CeO2-incorporated HA coatings reversed the decreased superoxide dismutase (SOD) activity, reduced reactive oxygen species (ROS) generation, and suppressed the malondiadehyde (MDA) formation. The findings suggested that CeO2-modified HA coatings may be promising coating materials for osteoporotic bone regeneration.

The Effects of Cerium Valence States at Cerium Oxide Coatings on the Responses of Bone Mesenchymal Stem Cells and Macrophages.

Ideal orthopedic coatings should trigger good osteogenic response and limited inflammatory response. The cerium valence states in ceria are associated with their anti-oxidative activity and anti-inflammatory property. In the study, we prepared two kinds of plasma sprayed CeO2 coatings with different Ce4+ concentrations to investigate the effects of Ce valence states on the response of bone mesenchymal stem cells (BMSCs) and macrophage RAW264.7. Both the coatings (CeO2-A and CeO2-B) were characterized via XRD, SEM, and X-ray photoelectron spectroscopy. The CeO2 coatings enhanced osteogenic behaviors of BMSCs in terms of cellular proliferation, alkaline phosphatase (ALP) activity and calcium deposition activity in comparison with the Ti substrate. In particular, the CeO2-B coating (higher Ce4+ concentration) elicited greater effects than the CeO2-A coating (higher Ce3+ concentration). RT-PCR and western blot results suggested that the CeO2-B coating promoted BMSCs osteogenic differentiation through the SMAD-dependent BMP signaling pathway, which activated Runx2 expression and subsequently enhanced the expression of ALP and OCN. With respect to either CeO2-A coating or Ti substrate, the CeO2-B coating exerted greater effects on the macrophages, increasing the anti-inflammatory cytokines (IL-10 and IL-1ra) expression and suppressing the expression of the pro-inflammatory cytokines (TNF-α and IL-6) and ROS production. Furthermore, it also upregulated the expression of osteoinductive molecules (TGF-ß1 and BMP2) in the macrophages. The regulation of cerium valence states at plasma sprayed ceria coatings can be a valuable strategy to improve osteogenic properties and alleviate inflammatory response.

Incorporation of cerium oxide into hydroxyapatite coating regulates osteogenic activity of mesenchymal stem cell and macrophage polarization.

Biomedical coatings for orthopedic implants should facilitate osseointegration and mitigate implant-induced inflammatory reactions. Cerium oxide (CeO2) ceramics possess anti-oxidative properties and can be used to decrease mediators of inflammation, which makes them attractive for biomedical applications. In our work, two kinds of CeO2 incorporated hydroxyapatite coatings (HA-10Ce and HA-30Ce) were prepared via plasma spraying technique and the effects of CeO2 addition on the responses of bone mesenchymal stem cells (BMSCs) and RAW264.7 macrophages were investigated. An increase in CeO2 content in the HA coatings resulted in better osteogenic behaviors of BMSCs in terms of cell proliferation, alkaline phosphatase (ALP) activity and mineralized nodule formation. RT-PCR and western blot analysis suggested that the incorporation of CeO2 may promote the osteogenic differentiation of BMSCs through the Smad-dependent BMP signaling pathway, which activated Runx2 expression and subsequently enhanced the expression of ALP and OCN. The expression profiles of macrophages cultured on the CeO2 modified coating revealed a tendency toward a M2 phenotype, because of an upregulation of M2 surface markers (CD163 and CD206), anti-inflammatory cytokines (TNF-α and IL-6) and osteoblastogenesis-related genes (BMP2 and TGF-ß1) as well as a downregulation of M1 surface markers (CCR7 and CD11c), proinflammatory cytokines (IL-10 and IL-1ra) and reactive oxygen species production. The results suggested the regulation of BMSCs behaviors and macrophage-mediated responses at the coating's surface were associated with CeO2 incorporation. The incorporation of CeO2 in HA coatings can be a valuable strategy to promote osteogenic responses and reduce inflammatory reactions.

The Effects of Cerium Oxide Incorporation in Calcium Silicate Coating on Bone Mesenchymal Stem Cell and Macrophage Responses.

Ideal coatings for orthopedic implants should be able to induce excellent osseointegration with host bone tissue, which requires good osteogenic responses and limited inflammatory reactions. Cerium oxide (CeO2) ceramics have anti-oxidative properties and can be used to decrease mediators of inflammation, making them attractive for biomedical application. In this study, two kinds of CeO2 incorporated calcium silicate coatings (CS-10Ce and CS-30Ce) were prepared via plasma spraying technique, and the effects of CeO2 addition on the responses of bone mesenchymal stem cells (BMSCs) and RAW264.7 macrophages were evaluated. The CS-10Ce and CS-30Ce coatings were characterized by X-ray diffraction, scanning electron microscopy, and X-ray photoelectron spectroscopy. An increase in CeO2 content in the coatings resulted in enhanced chemical stability and better BMSCs osteogenic behaviors in terms of cell adhesion, proliferation, ALP activity, and mineralized nodule formation. With respect to either ZrO2-added or unmodified CS coating, the CS-30Ce coating elicited higher effects on the macrophages, suppressing the gene expressions of pro-inflammatory (M1) markers (CCR7, IL-6, and TNF-α), while upregulating the expressions of anti-inflammatory (M2) markers (CD206, IL-1ra, and IL-10); moreover, it could also increase the expression of osteoinductive molecules (BMP2 and TGF-ß1) by the macrophages. The results suggested that the regulation of BMSCs behaviors and macrophage-mediated responses at the coating's surface was related to CeO2 incorporation. The incorporation of CeO2 in CS coatings can be a valuable strategy to promote osteogenic responses and mitigate inflammatory reactions.

Cerium Oxide-Incorporated Calcium Silicate Coating Protects MC3T3-E1 Osteoblastic Cells from H2O2-Induced Oxidative Stress.

Oxidative stress regulates cellular functions in multiple pathological conditions, including bone formation by osteoblastic cells. In this work, the protective effects of cerium oxide (CeO2)-incorporated calcium silicate (CeO2-CS) coating on the response of osteoblasts to H2O2-induced oxidative stress and the related mechanism were examined. CeO2 incorporation significantly improved osteoblast viability and reduced cell apoptosis caused by H2O2 when compared with the control. H2O2-induced reduction of differentiation marker alkaline phosphatase (ALP) was recovered in the presence of the CeO2-CS coating. The above effects were mediated by the antioxidant effect of CeO2. The CeO2-CS coating immersed in 0.1 mM H2O2 aqueous solution was able to degrade 64 % of it in 1 week. In addition, CeO2 incorporation decreased reactive oxygen species (ROS) production and suppressed malondialdehyde (MDA) formation in H2O2-treated osteoblasts. Taken together, CeO2-CS biomedical coatings with antioxidant property would be promising for bone regeneration under oxidative stress.

Plasma sprayed cerium oxide coating inhibits H2O2-induced oxidative stress and supports cell viability.

Oxidative stress is a risk factor in the pathogenesis of osteoporosis, and plays a major role in bone regeneration of osteoporotic patients. Cerium oxide (CeO2) ceramics have the unique ability to protect various types of cells from oxidative damage, making them attractive for biomedical applications. In this study, we developed a plasma sprayed CeO2 coating with a hierarchical topography where ceria nanoparticles were superimposed in the micro-rough coating surface. The protective effects of the CeO2 coating on the response of osteoblasts to H2O2-induced oxidative stress have been demonstrated in terms of cell viability, apoptosis and differentiation. The CeO2 coating reversed the reduced superoxide dismutase activity, decreased reactive oxygen species production and suppressed malondialdehyde formation in H2O2-treated osteoblasts. It indicated that the CeO2 coating can preserve the intracellular antioxidant defense system. The cytocompatibility of the CeO2 coating was further assessed in vitro by cell viability assay and scanning electron microscopy analysis. Taken together, the CeO2 coating could provide an opportunity to be utilized as a potential candidate for bone regeneration under oxidative stress.

Cough event classification by pretrained deep neural network.

BACKGROUND: Cough is an essential symptom in respiratory diseases. In the measurement of cough severity, an accurate and objective cough monitor is expected by respiratory disease society. This paper aims to introduce a better performed algorithm, pretrained deep neural network (DNN), to the cough classification problem, which is a key step in the cough monitor. METHOD: The deep neural network models are built from two steps, pretrain and fine-tuning, followed by a Hidden Markov Model (HMM) decoder to capture tamporal information of the audio signals. By unsupervised pretraining a deep belief network, a good initialization for a deep neural network is learned. Then the fine-tuning step is a back propogation tuning the neural network so that it can predict the observation probability associated with each HMM states, where the HMM states are originally achieved by force-alignment with a Gaussian Mixture Model Hidden Markov Model (GMM-HMM) on the training samples. Three cough HMMs and one noncough HMM are employed to model coughs and noncoughs respectively. The final decision is made based on viterbi decoding algorihtm that generates the most likely HMM sequence for each sample. A sample is labeled as cough if a cough HMM is found in the sequence. RESULTS: The experiments were conducted on a dataset that was collected from 22 patients with respiratory diseases. Patient dependent (PD) and patient independent (PI) experimental settings were used to evaluate the models. Five criteria, sensitivity, specificity, F1, macro average and micro average are shown to depict different aspects of the models. From overall evaluation criteria, the DNN based methods are superior to traditional GMM-HMM based method on F1 and micro average with maximal 14% and 11% error reduction in PD and 7% and 10% in PI, meanwhile keep similar performances on macro average. They also surpass GMM-HMM model on specificity with maximal 14% error reduction on both PD and PI. CONCLUSIONS: In this paper, we tried pretrained deep neural network in cough classification problem. Our results showed that comparing with the conventional GMM-HMM framework, the HMM-DNN could get better overall performance on cough classification task.

ISMAC: An Intelligent System for Customized Clinical Case Management and Analysis.

Clinical cases are primary and vital evidence for Traditional Chinese Medicine (TCM) clinical research. A great deal of medical knowledge is hidden in the clinical cases of the highly experienced TCM practitioner. With a deep Chinese culture background and years of clinical experience, an experienced TCM specialist usually has his or her unique clinical pattern and diagnosis idea. Preserving huge clinical cases of experienced TCM practitioners as well as exploring the inherent knowledge is then an important but arduous task. The novel system ISMAC (Intelligent System for Management and Analysis of Clinical Cases in TCM) is designed and implemented for customized management and intelligent analysis of TCM clinical data. Customized templates with standard and expert-standard symptoms, diseases, syndromes, and Chinese Medince Formula (CMF) are constructed in ISMAC, according to the clinical diagnosis and treatment characteristic of each TCM specialist. With these templates, clinical cases are archived in order to maintain their original characteristics. Varying data analysis and mining methods, grouped as Basic Analysis, Association Rule, Feature Reduction, Cluster, Pattern Classification, and Pattern Prediction, are implemented in the system. With a flexible dataset retrieval mechanism, ISMAC is a powerful and convenient system for clinical case analysis and clinical knowledge discovery.

Intelligent ZHENG Classification of Hypertension Depending on ML-kNN and Information Fusion.

Hypertension is one of the major causes of heart cerebrovascular diseases. With a good accumulation of hypertension clinical data on hand, research on hypertension's ZHENG differentiation is an important and attractive topic, as Traditional Chinese Medicine (TCM) lies primarily in "treatment based on ZHENG differentiation." From the view of data mining, ZHENG differentiation is modeled as a classification problem. In this paper, ML-kNN-a multilabel learning model-is used as the classification model for hypertension. Feature-level information fusion is also used for further utilization of all information. Experiment results show that ML-kNN can model the hypertension's ZHENG differentiation well. Information fusion helps improve models' performance.

Inquiry diagnosis of coronary heart disease in Chinese medicine based on symptom-syndrome interactions.

BACKGROUND: There is a long history of coronary heart disease (CHD) diagnosis and treatment in Chinese medicine (CM), but a formalized description of CM knowledge is still unavailable. This study aims to analyze a set of CM clinical data, which is important and urgent. METHODS: Relative associated density (RAD) was used to analyze the one-way links between the symptoms or syndromes or both. RAD results were further used in symptom selection. RESULTS: Analysis of a dataset of clinical CHD diagnosis revealed some significant relationships, not only between syndromes but also between symptoms and syndromes. Using RAD to select symptoms based on different classifiers improved the accuracy of syndrome prediction. Compared with other traditional symptom selection methods, RAD provided a higher interpretability of the CM data. CONCLUSION: The RAD method is effective for CM clinical data analysis, particular for analysis of relationships between symptoms in diagnosis and generation of compact and comprehensible symptom feature subsets.

MAPLSC: a novel multi-class classifier for medical diagnosis.

Analysis of clinical records contributes to the Traditional Chinese Medicine (TCM) experience expansion and techniques promotion. More than two diagnostic classes (diagnostic syndromes) in the clinical records raise a popular data mining problem: multi-value classification. In this paper, we propose a novel multi-class classifier, named Multiple Asymmetric Partial Least Squares Classifier (MAPLSC). MAPLSC attempts to be robust facing imbalanced data distribution in the multi-value classification. Elaborated comparisons with other seven state-of-the-art methods on two TCM clinical datasets and four public microarray datasets demonstrate MAPLSC's remarkable improvements.