Obstructive sleep apnea (OSA) is a common sleep disorder that is often associated with reduced heart rate variability (HRV), thus reflecting modulation of the autonomic system. Sliding trend fuzzy approximate entropy (SlTr-fApEn), which is based on the empirical mode decomposition (EMD) method, has been proposed as a novel index for analyzing HRV with OSA. This study included 60 electrocardiogram recordings from the PhysioNet database (40 OSA recordings and 20 healthy recordings) with apnea or no apnea in 5-minute segments. HRV indices obtained by sliding trend analysis were compared to those obtained by time-frequency domain analysis. Among all indices, the ratio of low-frequency power and high-frequency power (LF/HF) and sliding trend indices could significantly differentiate OSA recordings from normal recordings (p < 0.05). The OSA screening accuracy of SlTr-fApEn (85%) was higher than that of LF/HF (80%). Disease state analysis showed significant differences in SlTr-fApEn among the control group, normal OSA group, and apnea OSA group (p < 0.05). Therefore, SlTr-fApEn can reflect the complexity of autonomic changes during a short time period.

通过测定多种牙膏常用的抑菌药物的最低抑菌浓度(MIC),发现西吡氯铵(CPC)和氯己定对变异链球菌和牙龈卟啉单胞菌均具有强烈的抑制作用,其MIC均低于1 mg/L.利用三乙酰基-β-环糊精(TA-CD)和乙基纤维素(EC)对CPC进行包裹,喷雾干燥制样,得到的载药颗粒具有明显的"暴释"现象,加入硬脂酸(SA)后,载药颗粒表现出更好的缓释效果.另外,在载药颗粒中添加一定量的聚乙烯醇(PVA),可以明显提升载药颗粒刷牙后在口腔的附着率,当PVA的质量分数为20％时,载药颗粒的附着率最高,达到19.1％.利用混合菌生物膜对DLP-7载药颗粒和CPC进行抑菌评价,实验结果表明,随着时间延长,载药颗粒抑菌率不断增加,4 h后其抑菌率要明显高于CPC对于生物膜的抑菌率,即DLP-7载药颗粒体现出长效的抑菌效果.

Obstructive sleep apnea (OSA) is the most common sleep-related breathing disorder that potentially threatened people's cardiovascular system. As an alternative to polysomnography for OSA detection, ECG-based methods have been developed for several years. However, previous work is focused on feature engineering, which is highly dependent on the prior knowledge of human experts and maybe subjective. Moreover, feature engineering also highlights the prominent shortcoming of current learning algorithms that the features are unable to extracted and organized from the data. In this study, we proposed a method to detect OSA based on deep neural network and Hidden Markov model (HMM) using single-lead ECG signal. The method utilized sparse auto-encoder to learn features, which belongs to unsupervised learning that only requires unlabeled ECG signals. Two types classifiers (SVM and ANN) are used to classify the features extracted from the sparse auto-encoder. Considering the temporal dependency, HMM was adopted to improve the classification accuracy. Finally, a decision fusion method is adopted to improve the classification performance. About 85% classification accuracy is achieved in the per-segment OSA detection, and the sensitivity is up to 88.9%. Based on the results of per-segment OSA detection, we perfectly separate the OSA recording from normal with accuracy of 100%. Experimental results demonstrated that our proposed method is reliable for OSA detection.

A novel sparsity-based stochastic pooling which integrates the advantages of max-pooling, average-pooling and stochastic pooling is introduced. The proposed pooling is designed to balance the advantages and disadvantages of max-pooling and average-pooling by using the degree of sparsity of activations and a control function to obtain an optimized representative feature value ranging from average value to maximum value of a pooling region. The optimized representative feature value is employed for probability weights assignment of activations in normal distribution. The proposed pooling also adopts weighted random sampling with a reservoir for the sampling process to preserve the advantages of stochastic pooling. This proposed pooling is evaluated on several standard datasets in deep learning framework to compare with various classic pooling methods. Experimental results show that it has good performance on improving recognition accuracy. The influence of changes to the feature parameter on recognition accuracy is also investigated.

There are extensive studies investigating congestive heart failure (CHF) detection based on heart rate variability. Although a high level of accuracy has been achieved, its robustness under different conditions is not guaranteed. To improve the robustness, we applied sparse auto-encoder-based deep learning algorithm in CHF detection with RR intervals. A total data size of 30,592 (5-min RR interval) was obtained from 72 healthy persons and 44 CHF patients. The deep learning algorithm first extracts unsupervised features using a sparse auto-encoder from raw RR intervals, then constructs a deep neural network model with various hidden nodes combinations. Results showed that the model achieved 72.41% accuracy. This demonstrated that RR intervals have potential in CHF detection but cannot fully reflect dynamic change in 24-h.

Sleep posture has been used as a sleep assessment indicator in home health monitoring and clinical monitoring in recent years. Considering comfort and usability, unobtrusive sleep posture detection is needed. In this paper, we proposed a novel sleep respiration-derived posture (RDP) method based on left and right lung respiration impedance signals. We developed a dual-channel respiratory impedance acquisition system with wireless transmission. Then, support vector machine using radial basis function kernel was applied to recognize four typical sleep postures. Moreover, the performance of the SVM classifier was improved by using backward elimination. In-situ experiments with 16 subjects indicated that the RDP method reached an accuracy of 99.67%. Thus, our method is reliable in sleep posture recognition. Furthermore, a whole-night monitoring system based on respiration impedance can be conducted for sleep quality assessment.

Lung cancer bone metastases usually involve multiple interactions between cancer cells and host organs. To mimic such a structure, a cell co-culture micropattern of A549 lung cancer cells and osteoblast cells (A549/OB) was developed using a μ-eraser strategy for anti-cancer drug evaluation and also understanding cell–cell communication. When applying the μ-eraser strategy, a PDMS stamp was pressed to induce cell lysis. Hence, the minimum pressure required to induce cell lysis was first quantified. Then the pressure and pressing time for the micropatterning process were optimized to obtain cell micropatterns of high fidelity and repeatability. For different types of cells (A549 and human mesenchymal stem cells) and different substrates (TCP and PLGA nanofiber sheets), the optimized pressure was also different (i.e., for A549 on TCP, a pressure of 24.5 kPa was pressed for 10 s). According to a live/dead assay and Alamar Blue assay, cell viability and proliferation potential were not affected by the micropatterning process. There were several advantages for the μ-eraser strategy: substrates were not pre-treated; cell micropatterns mainly relied on the pattern of the PDMS stamp along with the micropatterning process; cells in the micropattern were not restricted in specified regions. Thus the A549/OB co-culture micropattern on TCP was used to evaluate the efficacy of a therapeutic agent (doxorubicin). In the co-culture micropattern, the efficacy of doxorubicin decreased when the expression of ALP in OB was elevated. The co-culture micropattern model showed the potential to be used for new anti-cancer drug development.

PAA modified Zn-doped HAp-like calcium phosphate (PAA-CaP/Zn) nanoparticles were homogeneously distributed in PLGA electrospun nanofibers, and enhanced the osteogenic differentiation of rADSCs.

Monitoring and timely intervention are extremely important in the continuous home care. Microneedle array electrode (MAE) have been employed for the long-term bio-signal monitoring without skin preparation. We developed a novel magneto-rheological drawing lithography (MRDL) method to cost-effectively fabricate a flexible micro-needle array electrode (FMAE) for the wearable bio-signal monitoring. Flexible substrate may match closely with curved skin and maintain a stable interface between skin and electrode. The formation mechanism of microneedle array (MA) by MRDL and bio-signal recording performance of FMAE were investigated. MA can be one-step drawn from the droplet array of curable magnetorheological fluid under the assist of external magnetic field. Ti/Au film was coated on the surface of solidified MA to insure the conductivity and compatibility of FMAE. 36-FMAE consists of 6 x 6 microneedles with an average height of 600 pm and an average tip radius of 12 mu m. FMAE with 36 needles (36-FMAE) shows a better bio-signal monitoring performance in some specific situations compared with flexible dry electrode (FDE) and commercial Ag/AgCl electrode. Electrode-skin interface impedance (EII) measured by 36-FMAE is the lowest at a given low input frequency and the amplitude of electrocardiography (ECG) and electroencephalography (EEG) signals recorded by 36-FMAE is the largest. 36-FMAE can collect more distinguishable features and weaken the effect of motion artifact during the dynamical ECG recording, Therefore, 36-FMAE is a promising sensor for the wearable bio-signal monitoring in home care. (C) 2017 Elsevier B.V. All rights reserved.

Graphene and its derivatives, along with adipose-derived stem cells (ADSCs), have demonstrated great potential to be used in neuron tissue engineering. ADSCs from rats (rADSCs) were first isolated and the surface markers were evaluated using flow cytometry. The rADSCs were fibroblast-like, positively expressed CD90 and CD105, lacked surface markers CD31, CD34 and CD45, and also showed adipogenic and osteogenic differentiation potentials. Graphene oxide (GO) and reduced graphene oxide (rGO) substrates were successfully fabricated. Based on X-ray photoelectron spectroscopy spectra, it was found that GO had more oxygen containing groups than rGO. On these substrates, rADSCs were spread; they proliferated very well and no cytotoxicity was observed. For GO, better organized actin in rADSCs was observed than that on rGO, which was probably due to the higher amount of oxygen in GO. After neurogenic induction, it was found that rADSCs transformed from a fibroblast-like to a spindle-like morphology when observed under a light microscope and scanning electron microscopy, and still no cytotoxic effect was found. Furthermore, immunofluorescence staining showed that neural-specific markers MAP2 and NeuN were detected in cells on these substrates. Both GO and rGO substrates were cytocompatible for rADSCs and rGO was probably more prominent in enhancing neurogenic differentiation.

In order to mimic the in vivo 3D micro environment and compare proliferation rate of adipose derived stem cells (ADSCs) on 2D and 3D aligned nanofiber sheets,3D aligned poly (caprolactone) (PCL) fiber scaffolds were prepared by improving the collector for electrospinning.The fibers were around capillaries in diameter of 300 μm,which were aligned very well and with fiber diameter in the range of 300 ～ 400 nm.After surface modification,ADSCs were seeded and it was found these cells were spread very well.Regardless 2D or 3D scaffolds,on aligned fibers cells were nearly all aligned along the fiber direction.When compared with that on 2D membranes,cells proliferated faster on 3D fiber scaffolds,showing the higher potential of 3D fiber scaffolds to be applied for tissue engineering applications.

We propose a method for human action recognition using latent topic models. There are two main differences between our method and previous latent topic models used for recognition problems. First, our model is trained in a supervised way, and we propose a two-level Beta process hidden Markov model which automatically identifies latent topics of action in video sequences. Second, we use the human skeleton to refine the spatial–temporal interest points that are extracted from video sequences. Because latent topics are derived from these interest points, the refined interest points can improve the precision of action recognition. Experimental results using the publicly available “Weizmann”, “KTH”, “UCF sport action”, “Hollywood2”, and “HMDB51” datasets demonstrate that our method outperforms other state-of-the-art methods.

For successful gene therapy, it is imperative to accumulate therapeutic gene in tumor tissues followed by efficiently delivering gene into targeted cells. Ultrasound irradiation, as a noninvasive and cost-effective external stimulus, has been proved to be one of the most potential external-stimulating gene delivery strategies recently in further improving gene transfection. In this study, we developed tumor-targeting ultrasound-triggered phase-transition nanodroplets AHNP-PFP-TNDs comprising a perfluorinated poly(amino acid) C11F17-PAsp (DET) as a core for simultaneously loading perfluoropentane (PFP) and nucleic acids, and a polyanionic polymer PGA-g-PEG-AHNP as the shell for not only modifying the surface of nanodroplets but also introducing an anti-Her2/neu peptide (AHNP) aiming to targeted treatment of Her2-overexpressing breast cancer. The results showed the average diameter of AHNP-PFP-TNDs was below 400 nm, nearly spherical in shape. The modification of PGA-g-PEG-AHNP not only increased the serum stability of the nanodroplets but also improved the affinity between nanodroplets and Her2-overexpressing breast cells. Both intratumor and intravenous injection of AHNP-PFP-TNDs into nude mice bearing HGC-27 xenografts showed that the gene transfection efficiency and the ultrasound contrast effect were significantly enhanced after exposed to the ultrasound irradiation with optimized ultrasound parameters. Therefore, this targeting nanodroplets system could be served as a potential theranostic vector for tumor targeting ultrasound diagnosis and gene therapy.

Background: Pectus excavatum (PE) is the most common chest wall malformation. The main treatments of PE are the Nuss and Ravitch procedures. Because the pathogenesis of PE remains unclear, an exact animal model has been difficult to establish. Animal models are unavailable to test new steel bars used in the Nuss procedure and to elucidate the mechanism by which PE affects pulmonary function. This study described the establishment of a rabbit model of PE. Methods: Twenty-four New Zealand white rabbits were randomized into a PE group and a control group. The animals in the PE group underwent surgery to remove a 0.5 cm long segment of the fifth to seventh costicartilage and removal of the sternum at the fifth costicartilage level. In control animals, the skin and muscle were incised and stitched into two layers. Results: Ten days after surgery, rabbits in the PE group showed gradual depression of the sternum. Over time, as rabbit weight increased, the depression of the anterior and inferior chest wall deepened and widened gradually. The deformity of the chest wall was similar at 6 and 12 weeks after surgery. Chest CT scans at 8 weeks after surgery showed that the depression extended from the cutting at the fifth costicartilage level to the spine. Three-dimensional reconstruction of the thorax showed that the depression of the sternum began at the level of fifth rib and most obvious at the level of the seventh rib. Conclusions: This rabbit model of PE was simple, less invasive, and easy to establish.

Saliva contains important personal physiological information that is related to some diseases, and it is a valuable source of biochemical information that can be collected rapidly, frequently, and without stress. In this article, we reported a new and simple localized surface plasmon resonance (LSPR) substrate composed of polyaniline (PANI)-gold hybrid nanostructures as an optical sensor for monitoring the pH of saliva samples. The overall appearance and topography of the substrates, the composition, and the wettability of the LSPR surfaces were characterized by optical and scanning electron microscope (SEM) images, infrared spectra, and contact angles measurement, respectively. The PANI-gold hybrid substrate readily responded to the pH. The response time was very short, which was 3.5 s when the pH switched from 2 to 7, and 4.5 s from 7 to 2. The changes of visible-near-infrared (NIR) spectra of this sensor upon varying pH in solution showed that—for the absorption at given wavelengths of 665 nm and 785 nm—the sensitivities were 0.0299 a.u./pH (a.u. = arbitrary unit) with a linear range of pH = 5–8 and 0.0234 a.u./pH with linear range of pH = 2–8, respectively. By using this new sensor, the pH of a real saliva sample was monitored and was consistent with the parallel measurements with a standard laboratory method. The results suggest that this novel LSPR sensor shows great potential in the field of mobile healthcare and home medical devices, and could also be modified by different sensitive materials to detect various molecules or ions in the future.

Using diagnostic imaging, asymmetry in the mechanic of the left and right lungs has been discovered in unilateral lung disease. Unilateral lung disease might affect information interaction between two lungs, which is often neglected in the study of lung function. In this paper, twenty-one subjects were recruited to collect bio-impedance respiratory signals of the left and right lungs. The differences of correlation and amplitude between the two respiratory signals were combined to verify three types of respiration movement: respiration symmetry (RS), respiration asymmetry type I (T1RA), and respiration asymmetry type II (T2RA). We extracted the correlation coefficient, mutual information (MI), and transfer entropy (TE) to evaluate lung function in these three types of lung respiration. Results showed that MI was significantly larger in RS than in T1RA and T2RA. TE in both directions (TE(L→R) and TE(R→L)) were significantly different in RS, T1RA, and T2RA. TE(R→L) increased progressively when shifting RS to T1RA and then to T2RA. The prominent direction of transferred information in two lungs was significantly different for T1RA and T2RA. The results indicate that MI is suitable parameters for detecting respiration symmetry and asymmetry. TE is a useful tool for detecting two respiration movement asymmetry. The results provide a better understanding of the mechanisms underlying responsible for pulmonary ventilation redistribution and could provide novel clinical markers to evaluate asymmetry of two lungs effectively.

To be better used as implant materials in bone regeneration, bioactivity and osteogenesis of poly(methyl methacrylate) (PMMA)-based bone cement need to be further enhanced. In this study, bioactive hydroxyapatite (HA) nanoparticle decorated with polydopamine (pDA) and BMP2 mimicking peptide (serine-serine-valine-proline-threonine, SSVPT) was incorporated into PMMA-based cement (p(x)-HA/PMMA). The presence of pDA layer on HA surface could keep the mechanical property of p(x)-HA/PMMA cements comply with the requirement of clinical application. More importantly, the p(x)-HA/PMMA cement exhibited excellent biomechanical property with enhanced interfacial adhesion between cement and swine femur. Biomineralization and cell experiment in vitro showed that the incorporation of pDA-coated HA and SSVPT can synergistically enhance the bioactivity of px-HA/PMMA cement for regeneration and provide an attractive strategy to reconstruction of load-bearing bone. (C) 2016 Elsevier Ltd. All rights reserved.

In this paper, a series of copolymer hydrogels were fabricated from methacrylated poly(γ-glutamic acid) (mPGA) and poly(ethylene glycol) diacrylate (PEGDA). The effect of ionic strength and pH on the swelling behavior and mechanical properties of these hydrogels were studied in detail. Release of Rhodamine B as a model drug from the hydrogel was evaluated under varied pH. In vitro photoencapsulation of bovine cartilage chondrocytes was performed to assess the cytotoxicity of this copolymer hydrogel. The results revealed that the copolymer hydrogel is ionic- and pH-sensitive, and does not exhibit acute cytotoxicity; this copolymer hydrogel may have promising application as matrix for controlled drug release and scaffolding material in tissue engineering.

Poly(lactic acid) stereocomplex (sc-PLA) prevails over homo-poly(lactic acid)s in many aspects and showed great potential as biomaterials depending on its extraordinary biological, thermal, and mechanical properties. In the past few years, sc-PLA has gained tremendous attention as a new approach to the development of new types of functional biomaterials. This paper summarizes recent progress in the application of the concept of sc-PLA in the design and fabrication of stabilized nanoparticles and in situ gelling hydrogels for biomedical applications.

A novel micro-needle array electrode (MAE) fabricated by thermal drawing and coated with Ti/Au film was proposed for bio-signals monitoring. A simple and effective setup was employed to form glassy-state poly (lactic-co-glycolic acid) (PLGA) into a micro-needle array (MA) by the thermal drawing method. The MA was composed of 6 × 6 micro-needles with an average height of about 500 μm. Electrode-skin interface impedance (EII) was recorded as the insertion force was applied on the MAE. The insertion process of the MAE was also simulated by the finite element method. Results showed that MAE could insert into skin with a relatively low compression force and maintain stable contact impedance between the MAE and skin. Bio-signals, including electromyography (EMG), electrocardiography (ECG), and electroencephalograph (EEG) were also collected. Test results showed that the MAE could record EMG, ECG, and EEG signals with good fidelity in shape and amplitude in comparison with the commercial Ag/AgCl electrodes, which proves that MAE is an alternative electrode for bio-signals monitoring.