ABSTRACT
In the era of big data, hospital′s research data platform development is faced with challenges in how to strengthen the deep and efficient application of medical big data in their clinical research. The authors studied such problems of a tertiary hospital in Henan as low data value density, difficulty in structuring medical text language and mismatch between clinical research thinking and ability, in terms of the use of existing clinical data exploration system and effective use of data. Based on results of the study, they summarized relevant measures for effective use of medical data and considerations in the establishment of a clinical research big data center.For example, combination of " pre-structuring" and " post-structuring" in data collection for data quality control; exploration of deep mining of medical text data using machine learning technology based on unstructured text data, and perfection the intelligent analysis application function of clinical research big data; enhancement of talent training to promote clinicians′ capacity in using clinical big data; and enhancement of multi-disciplinary teambuilding and composite talent cultivation, for the purposes of more efficient use of medical big data and higher efficiency and quality of clinical research.
ABSTRACT
Objective To observe the condition changes in patients with chronic hepatitis B (CHB) following medication withdraw of nucleoside/nucleotide analogues (NAs) treatment and to analyze the factors related to disease relapse.Methods Eighty-five CHB patients who discontinued medication of nucleoside/nucleotide analogues for antiviral therapy in Nanjing Second Municipal Hospital from January 2002 to December 2017 were enrolled in the study, among whom 22 cases met the withdrawal criteria (standard withdrawal group ) and 63 cases did not meet the withdraw criteria ( non-standard withdrawal group).The correlation of condition changes (abnormal liver function, positive rate of HBV DNA, hepatic failure with the drug withdrawal , the course of medication, serological transformation of HBeAg during drug withdrawal, HBsAg level, and liver cirrhosis during drug administration was analyzed.Results In standard withdraw group, the medication lasted for ＞3 years, only 1 case had HBV DNA positive conversion , abnormal ALT and TBil, and liver failure.In non-standard withdraw group, 50 cases (79.4%) had HBV DNA positive conversion, 36 (57.1%) had abnormal ALT, 25 ( 39.7%) had abnormal TBil and 14 (22.2%) had liver failure.There were 19 cases with HBsAg ＞1 000 IU/mL and 3 cases with HBsAg ≤1 000 IU/mL, and 1 case with HBsAg＞1 000 IU/mL (5.3%) had HBV DNA positive conversion , ALT, TBil abnormality and liver failure.In non-standard withdraw group, there were 52 cases with HBsAg ＞1000 IU/mL, among whom 45 cases (86.5%) had positive HBV DNA conversion , 31 (59.6%) had ALT abnormalities, 25 (48.1%) had TBil abnormalities, and 13 (25.0%) had liver failure; there were 11 patients with HBsAg ≤1 000 IU/mL, among whom 6 cases (54.5%) had HBV DNA positive conversion , 5 (45.4%) had ALT abnormalities, and no TBil abnormalities or liver failure occurred.There were 5 cases of liver cirrhosis in the standard withdraw group , only 1 case had HBV DNA positive conversion , ALT, TBil abnormality and liver failure.None of the 17 patients with non-cirrhosis had HBV DNA positive conversion , ALT, TBil abnormality and liver failure.There were 29 patients with liver cirrhosis in non-standard withdraw group showed positive HBV DNA conversion , 28 (96.6%) had ALT abnormalities, 22 (75.8%) had TBil abnormalities, and 11 (37.9%) had liver failure; among 34 non-cirrhosis patients, 21 (61.8%) had positive HBV DNA conversion, 8 (23.5%) had ALT abnormalities, 3 (8.8%) had TBil abnormalities, and 2 (5.9%) had liver failure.According to the standard discontinuation , 12 patients (16.7%) had positive HBV DNA transformation after HBeAg serological conversion , and no ALT abnormality, TBil abnormality and liver failure occurred.In non-standard withdraw group, only 17 cases without HBeAg serological conversion , 10 cases (58.8%) had positive HBV DNA conversion , 5 cases (29.4%) had ALT abnormalities, 2 cases (11.8%) had TBil abnormalities and liver failure did not occur.Conclusion CHB patients with medication of NAs should be discontinued according to the withdrawal criteria .and the course of medication, the immune index and the liver cirrhosis should be taken into account.
ABSTRACT
Objective To evaluate the relationship and screening value of percentage of body fat (BF%) and waist height ratio (WHtR) for hyperlipidemia in physical examination people. Methods A total of 2 668 objects taking physical examination in Henan Province People′s Hospital from September to December 2014 were included in this study. Values of height, body mass index (BMI), waist circumference, body composition and blood lipid level were detected. The receiver oper?ating characteristic curve (ROC) was used to analyze the screening rate of WHtR and BF%on hyperlipidemia with sensitivi?ty, specific and area under the curve (AUC). After stratified by age, waist circumference and BMI, multivariable logistic re?gression analysis was used to investigate the association between hyperlipidemia risk, BF% and WHtR. Results The screening accuracy rate on hyperlipidemia was higher for BF%, AUC was 0.79 in both female and male people. Among wom?en with BMI<18.5 kg/m2 and 18.5~<24 kg/m2, the risk of hyperlipidemia was higher in superfatted group than that of normal group. There was no correlation between WHtR and hyperlipidemia. Among men older than 40 y or with abnormal waist cir?cumference (≥85 cm), the risk of hyperlipidemia was higher in superfatted group than that of normal group, but not associat?ed with WHtR. Conclusion The BF%is a better screening indicator for hyperlipidemia compared with that of WHtR and BMI. Women with BMI<18.5 kg/m2 and 18.5~<24 kg/m2 and men older than 40 y or with waist circumference over 85 cm are suggested to do body composition tests to improve screening accuracy for hyperlipidemia.