Electrocardiogram signal classification algorithm of nested long short-term memory network based on focal loss function / 生物医学工程学杂志

Electrocardiogram (ECG) can visually reflect the physiological electrical activity of human heart, which is important in the field of arrhythmia detection and classification. To address the negative effect of label imbalance in ECG data on arrhythmia classification, this paper proposes a nested long short-term memory network (NLSTM) model for unbalanced ECG signal classification. The NLSTM is built to learn and memorize the temporal characteristics in complex signals, and the focal loss function is used to reduce the weights of easily identifiable samples. Then the residual attention mechanism is used to modify the assigned weights according to the importance of sample characteristic to solve the sample imbalance problem. Then the synthetic minority over-sampling technique is used to perform a simple manual oversampling process on the Massachusetts institute of technology and Beth Israel hospital arrhythmia (MIT-BIH-AR) database to further increase the classification accuracy of the model. Finally, the MIT-BIH arrhythmia database is applied to experimentally verify the above algorithms. The experimental results show that the proposed method can effectively solve the issues of imbalanced samples and unremarkable features in ECG signals, and the overall accuracy of the model reaches 98.34%. It also significantly improves the recognition and classification of minority samples and has provided a new feasible method for ECG-assisted diagnosis, which has practical application significance.

Research on high-efficiency electrocardiogram automatic classification based on autoregressive moving average model fitting / 生物医学工程学杂志

The automatic detection of arrhythmia is of great significance for the early prevention and diagnosis of cardiovascular diseases. Traditional arrhythmia diagnosis is limited by expert knowledge and complex algorithms, and lacks multi-dimensional feature representation capabilities, which is not suitable for wearable electrocardiogram (ECG) monitoring equipment. This study proposed a feature extraction method based on autoregressive moving average (ARMA) model fitting. Different types of heartbeats were used as model inputs, and the characteristic of fast and smooth signal was used to select the appropriate order for the arrhythmia signal to perform coefficient fitting, and complete the ECG feature extraction. The feature vectors were input to the support vector machine (SVM) classifier and K-nearest neighbor classifier (KNN) for automatic ECG classification. MIT-BIH arrhythmia database and MIT-BIH atrial fibrillation database were used to verify in the experiment. The experimental results showed that the feature engineering composed of the fitting coefficients of the ARMA model combined with the SVM classifier obtained a recall rate of 98.2% and a precision rate of 98.4%, and the

Origin and Phylogenetic Characteristics of Dengue Virus Isolated from the Large Outbreak of Dengue in Guangdong Province in 2014 / 中山大学学报(医学科学版)

[Objects]To isolate and identify the pathogen of the large outbreak of dengue in Guangdong province in 2014. To understand the origin and the phylogenetic characteristics of the isolates ,and provide scientific foundation for the surveillance and prevention of dengue fever.[Methods]Collected the patient serum samples over all the Guangdong province during the 2014 outbreakperiod,isolated and identified the virus from these samples. Amplified complete E gene and complete genome with certain primers and sequenced all the products. Then the Phylogenetic ,Bayesian phylogeography and mutations analysis were carried.[Results]40 DENV-1 strains were isolated and identified. 40 complete E gene sequences and 6 complete genome sequences of DENV-1 were obtained. Phylogenetic analysis with E gene sequences revealed that the 40 isolates were classified into two genotypes including 16 genotypeⅠ(Asia)and 24 genotypeⅤ(America/Africa). 14 genotypeⅠisolates were clustered closest with isolates from Guangdong province(2013)and Sigapore(2013)which share the nucletide identities of 99.6% ~ 99.9%,other two genotypeⅠisolates were clustered with strains from Malaysia (2013) and both share the nucletide identities of 99.7%;24 genotypeⅤisolates were all classified in one clade with striains from Bangladesh(2009),China(2009)and Bhutan(2013)which share nucletide identities of 99.0%-99.9%. Further analysis with six complete genome sequences showed that five isolates were clustered closest with strains isolated from Guangdong province(2013)share the nucletide identities of 99.6%-99.8% while the sixth stains closest with strains isolated from Myanmar(2002)share the nucletide identities of 98.8%. The isolates have five amino acid mutations compared with strains epidemic in Guangdong province in 2013,three mutations(S88V,E203G,T275R)are in the EⅡdomain and one mutation (S305P)is in the EⅢdomain which associated with virulence.[Conclusions]During the outbreak in Guangdong province in 2014, DENV-1 is the predominant causative serotype,and there are at least two different kinds of genotypes of DENV-1 largely epidemiced in the whole province. Evolution analysis reveals the multiple origins of the isolates which may origin from Guangdong province , Sigapore,Malaysia,Myanmar so that we should enhance the study and surveillance of autochthonous and vectors in order to understand the epidemic way of dengue in Guangdong province. The isolates have had four mutations in the domain associated with virulence which remain further study to know their biological effects.

Optimization of biotransformation of ursolic acid by alternaria longipes / 中国中药杂志

<p><b>OBJECTIVE</b>To conduct the biotransformation of ursolic acid by alternaria longipes AS3. 2875, and optimize the culture medium and biotransformation conditions.</p><p><b>METHOD</b>With the consumption rate of ursolic acid and the generation rate of 28-O-beta-D-glucopyranose ursolic acid as indicators, the impact of different biotransformation conditions such as pH, phosphate, different kinds of metal ions, spore concentration, substrate quantity, temperature, shaking speed and cultivation time on the transformation of ursolic acid in alternaria longipes culture solution were detected to obtain the optimal biotransformation conditions of ursolic acid by alternaria longipes.</p><p><b>RESULT</b>The optimized biotransformation conditions were as follows: initial pH value was 5.0, MgSO4 was 0.25 g x L(-1) K2HPO4 was 1.0 g x L(-1), FeSO4 was 0.083 g x L(-1), spore concentration was 4%, substrate quantity was 0.3 g x L(-1), shaking speed was 140 r x min(-1), cultivation temperature was at 28 degrees C and cultivating time was 3 days.</p><p><b>CONCLUSION</b>The generation rate of 28-O-beta-D-glucopyranose ursolic acid by alternaria longipes stabilizes at around 5%.</p>

Genetic Evolution of Neuraminidase Gene of Influenza A/H1N1 Virus / 中山大学学报(医学科学版)

[Objective]This study was designed to investigate the genetic evolution of the neuraminidase(NA)gene of seasonal A/H1N1 and 2009 novel A/H1N1 inflilenza virus,and discuss the genetic variation of influenza A virus.[Methods]The virus strains were separately isolated from the clinical samples collected in 2006 and 2009,and then identified as seasonal A/H1N1 and novel A/H1N1.The full length of the NA gene of these strains was amplified by RT-PCR.Then the genetic evolution and mutations of important functional sites were analyzed.[Results]The homology of NA gene between the 2009 novel A/H1N1 isolates and 2006 seasonal A/H1N1 isolates was low(77.9%～78.8%),so was the homology of NA gene between the 2009 novel A/H1N1 isolates and representative strains of different periods and 1979-2001 WHO recommended vaccine strains(78.1%～79.3%).But compared with the WHO recommended vaccine strains of 2009 novel A/H1N1,the homology reached more than 99%.The genetic evolution analysis revealed that NA gene of 2009 novel A/H1N1 had the closest genetic relationship with the swine influenza A virus(A/swine/Belgium/1/1983)from Eurasian Iineage,and some of the antigenic sites and neuraminidase active sites of NA gene of seasonal A/H1N1 were mutated after 2005.[Conclusion]The NA gene of 2009 novel A/H1N1 may originate from Eurasian Iineage of swine influenza virus.The variation of NA gene of seasonal A/H1N1 has occurred in a certain degree.Hence,it is very necessary to continuously monitor the variant of influenza A virus.

Verification of colon-targeted release of Changankang Pellets / 中成药

Objective:To determine the in vitro release of two kinds of Changankang Pellets (Radix Astragall, Rhizama coptid) coated with Surelease and Eudragit respectively and verify their colon-targeted release in vivo through X-ray photography. Methods: Changankang Pellets were made by mixing the medical BaSO 4 powder and appropriate drugs together according to original produce and coat methods. After determining their in vitro release, they were tested in vivo by 2 volunteers. X-ray films were taken after appropriate time and analyzed to determine the location and status of these pellets inside healthy objects. Results: Pellets made by this way can carry drugs through stomach, intestine and targeted-release them in colon; the coating combined with pH-dependent and time-dependent Eudragit thin layer has better results. Conclusion: X-ray radiography applying BaSO 4 powder could be taken as a quality control method for colon-targeted release drugs. The preliminary results showed that Changankang Pellets met the requirements of colon-targeted release.