The method for detecting the negative terms in Chinese electronic medical record (EMR) is useful in providing evidence for constructing concept index. In this respect, we adopted an improved method which combined maximum matching with mutual information in order to extract terms in EMRs. This method can overcome the influence of overlay ambiguity. In addition, for the determination of negative semantic, we also adopted an improved method which combined rule-based method with word co-occurrence. This new method can reduce the probability of appearance of false positive terms caused by punctuation input errors. The result showed that the negative predictive value is 7.85% higher than the rule-based method.
China ; Electronic Health Records ; Probability ; Semantics ; Terminology as Topic

Objective:To observe the changes of blood flow density and perfusion density in the macula of non-diabetic peritoneal dialysis (PD) patients, and their correlation with blood pressure, total protein, albumin, prealbumin, serum creatinine, urea, and high-sensitivity C-reactive protein were preliminarily analyzed.Methods:A single-center, cross-sectional, clinical observational study. From January to December 2018, 63 eyes of 63 non-diabetic patients (non-diabetic PD group) and 75 eyes of normal healthy people (the normal control group) who underwent PD treatment at the PD Center of Peking University First Hospital were included in the study. All were monocular into the group. Among the 63 patients in the non-diabetic PD group, 24 were males and 39 were females. The duration of PD was 7 to 185 months, with the average duration of 67.87±48.36 months. There were 75 healthy persons in the normal control group. There was no significant difference in age ( t=-0.558), sex ratio ( χ2=0.492), axial length ( t=-1.197), and BCVA between the two groups ( P>0.05). OCT angiography was used to scan the macular area of 3 mm×3 mm and 6 mm×6 mm in the subject’s right eye. The blood flow density and perfusion density of superficial retinal capillaries in the macular area, as well as the area, circumference, and morphological index of the foveal avascular zone (FAZ) were measured. The blood flow density and perfusion density at different locations in the macular area of the two groups of eyes were compared by independent sample t test. The blood pressure, total protein, albumin, prealbumin, serum creatinine, urea, and high-sensitivity C-reactive protein was performed by Pearson correlation analysis. Results:Compared with the healthy control group, the blood flow density and perfusion density of superficial retinal capillaries in the macular area of the non-diabetic PD group decreased in different scanning ranges with the macular vessel 3×3 center ( t=-2.409), the macular vessel 3×3 macular ( t=-2.423), macular vessel 3×3 intact ( t=-2.759), macular vessel 6×6 intact ( t=-1.882), macular vessel 6×6 outer layer ( t=-2.188), macular perfusion 3×3 center ( t=-1.990), macular perfusion 3×3 complete ( t=-2.719), macular perfusion 6×6 complete ( t=-2.113), and macular perfusion 6×6 outer layer ( t=-2.205). The difference was statistically significant ( P<0.05). The comparison of the macular FAZ area of the two groups of eyes was statistically significant ( t=1.985, P <0.05). Correlation analysis showed that 3×3 macular blood vessels were intact and mean arterial pressure was positively correlated ( r=0.256, P=0.043). The macular blood vessels were 3×3 intact, macular perfusion was 3×3 intact, and macular blood was 6×6 intact, which the pre-white protein was positively correlated with ( r=0.468, 0.362, 0.333; P<0.001, P=0.004, 0.008). The macular vessel 3×3 was intact, the macular perfusion 6×6 was intact, which the hypersensitive C-reactive protein was negatively correlated with ( r=-0.370, -0.287, P=0.005, 0.030). Conclusion:The superficial retinal blood flow density and perfusion density in the macular area of non-diabetic PD patients are lower than those of normal healthy people.

ObjectiveTo study the mechanism of Mahuang Xixin Fuzitang （MXFT） against migraine. MethodTraditional Chinese Medicine Systems Pharmacology Database and Analysis Platform （TCMSP）， SiwssTargetPrediction and other databases were used to screen the active components and action targets of MXFT as well as migraine-related targets. The potential protein-protein interaction （PPI） network diagram was plotted for the intersection targets of MXFT and migraine using STRING 11.5. Metascape was used for Gene Ontology （GO） and Kyoto Encyclopedia of Genes and Genomes （KEGG） enrichment analysis of potential intersection targets. The component-target-pathway network of MXFT was constructed by Cytoscape 3.7.1 to screen core targets with high degree value. Finally， the binding strength between core target and its mapping components was verified by molecular docking， and the core targets with desirable docking results were verified by animal experiments in vivo. Forty eight SD rats were selected， and except the blank group， the other rats were subcutaneously injected with nitroglycerin to prepare the migraine rat model. The modeled rats were randomly divided into model group， positive drug group and MXFT high-， medium- and low-dose groups. The positive drug group was given zolmitriptan tablets， and the MXFT high-， medium- and low-dose groups were given high， medium and low doses of MXFT， respectively. The changes of behavior and pain threshold of rats in each group were observed every other day after modeling. The levels of calcitonin gene-related peptide （CGRP）， extracellular signal-regulated kinase 2 （ERK2） and c-fos proto-oncogene （FOS） protein in plasma were detected by enzyme-linked immunosorbent assay （ELISA）. Immunohistochemical technique and Western blot were employed to determine the levels of extracellular signal-regulated kinase 1/2 （ERK1/2， also known as MAPK1/3） and protein kinase B 1 （Akt1）， protein kinase C α （PRKCA） in trigeminal nerve of SD rats. ResultThe network pharmacology showed that the core targets of MXFT in the treatment of migraine were MAPK1， MAPK3， Akt1， PRKCA， etc.， mainly involving neuroactive ligand-receptor interaction signaling pathway， calcium signaling pathway， MAPK signaling pathway， etc. The molecular docking demonstrated that MAPK1， MAPK3， Akt1， PRKCA， PRKCB and PRKCG had good binding ability with their mapping components. The animal experiments indicated that compared with the conditions in the blank group， the number of head scratching in the model group was increased （P<0.01）， and the pain threshold was decreased （P<0.01）. Compared with the conditions in the model group， the number of head scratching in each administration group was reduced （P<0.01）， and the pain threshold was increased （P<0.01）. In addition， the levels of CGRP， ERK2 and FOS proteins in plasma， and Akt1， ERK1/2 and PRKCA proteins in trigeminal ganglion of the model group were higher than those of the blank group （P<0.05， P<0.01）. The levels of CGRP， ERK2 and FOS proteins in plasma and Akt1， ERK1/2 and PRKCA proteins in trigeminal ganglion of each administration group were lower than those of the model group （P<0.05， P<0.01）. ConclusionMXFT had multi-component， multi-target and multi-pathway characteristics in the treatment of migraine， and the mechanism might be related to inhibiting vasodilation， reducing the release of inflammatory factors and inhibiting neuronal hyperactivity.