ABSTRACT
Objective:To explore the predictive value of artificial intelligence algorithm model for venous thromboembolism（VTE）in lower extremities of trauma patients.Methods:The data of 15，856 orthopedic inpatients were retrospectively collected from the information system database in Chinese PLA General Hospital from December 1992 to November 2017. The patients were divided according to whether they had thrombosis or not. Data pretreatment and feature extraction were carried out. Four artificial intelligence algorithms including Random Forest（RF），Bayes（Bayes），Decision Tree（DTC）and Gradient Boosting Tree（GBDT）were constructed to evaluate their clinical diagnostic efficacy in VTE. The original data were divided into training set and test set according to the ratio of 8∶2 by random stratified sampling method. By comparing the area under receiver operating characteristic curve（ROC）（AUC），true positive rate（TPR）and accuracy in the above methods，the efficiency of different models in clinical diagnosis of VTE was evaluated. According to the contribution degree of the features in the model，the important features were ranked to screen the independent risk factors of VTE.Results:For RF，Bayes，DTC and GBDT algorithm models，the AUC was 0.89，0.86，0.68，0.71，with the TPR for 0.29，0.44，0.38，0.66 and the accuracy for 0.97，0.94，0.95，0.76，respectively. The RF algorithm model had the highest accuracy and the largest AUC. Analysis of important features of artificial intelligence prediction models for VTE showed that the history of thrombosis was the primary predictor of adverse outcomes. The ranking of important clinical features represented by the RF model showed that the history of thrombosis，enoxaparin sodium injection dose，last glucose measurement and first glucose measurement after surgery were important predictive characteristics of VTE.Conclusions:The RF model has the highest accuracy in risk prediction of VTE in trauma patients，which can provide a reference for the formulation of VTE prevention strategies.
ABSTRACT
Objective To analyze primary diseases and risk factors of chronic cough in children,and develop clinical thinking for the doctor,looking for the orderly diagnosis method.Methods The clinical data of 123 children with chronic cough(medical history,physical examination,routine chest X-ray,PPD test,mycoplasma,Chlamydia antibody,antibody of respiratory syncytial virus,adenovirus IgM determination of IgM determination,Coxsackie virus IgM determination,when necessary,be lung CT,CT of paranasal sinuses,gastrointestinal barium meal,bronchiectasis agent diagnostic treatment and surgery consultation) were retrospectively analyzed.The cause of chronic cough in children with primary disease and related factors were analyzed.Results 123 cases of chronic cough in children's primary diseases were asthma-related cough in 57 cases (46.3 %),upper airway cough syndrome (rhinitis,allergic rhinitis,sinusitis) in 41 cases (33.3 %),chronic pharyngitis and tonsillitis,bronchitis in 27 cases (22.0%) ; the main etiology for Mycoplasma Chlamydia(48.8%,60 cases adenovirus),19 cases(15.4%),12 cases of respiratory syncytial virus (9.8%).Conclusion The main primary disease cough,asthma associated upper airway cough syndrome,chronic pharyngitis and tonsillitis,bronchitis,chronic cough in children,the main pathogen Chlamydia,Mycoplasma-for adenovirus,respiratory syncytial virus infection,diagnosis should be based on detailed,comprehensive medical history and physical examination,from simple to complex,according to from low to high,from conventional to special,from noninvasive to invasive principles are examined.
ABSTRACT
Objeetive To evaluate the performance characteristics of the second trimester double test for the detection of fetal Down's syndrome(DS)in women of advanced maternal age(AMA).Methods We undertook a prospective nation-wide multi-centered study and chose alpha-fetoprotein(AFP))and free β-subunit of human chorionic gonadotrophin(free β-hCG)as the serum markers.Between May 2004 and September 2006,12 centers participated in the collection and analysis of maternal serum AFP and free β-hCG.Patients with an iuereaged risk of DS(≥1/270)wero offered generic sunniocentesis.Follow up of the outcome of all pregnancies was obtained.Patients were divided into two groups,the AMA group and the non-AMA group and the screening efficiency Was evaluated in beth groups.Results A total of 66 132 singleton pregnancies were included in the study.and there were 36 10(5.46%)AMA women.The median maternal age of AMA women was 36.8years(35-47 years).At a cut-off of 1/270,in the AMA group,the number of positive cases screened was 727 and 22 cases of fetal DS were detected:the number of negative cases screened was 2883,and no fetal DS was found.In the non-AMA group,the number of positive cases screened was 4743 and 69 cases of fetal DS were detected:the number of negative cases screened was 57 779,and 6 cases of fetal DS were diagnosed postnatally.In AMA group,the detection rate(DR),false positive rate(FPR)and odds of being affected given a positive result(OAPR)were 100%,19.7%and 3.0%respectively.In the non-AMA group,the DR,FPR and OAPR were 92.0%.7.5%and 1.5%respectively.Conclusion The double-marker test using AFP and free β-hCG is an effective screen strategy for second-trimester detection of Down syndrome in AMA women.
ABSTRACT
Objective To evaluate the performance characteristics of the second trimester double-marker test for the detection of fetal Down's syndrome in mainland China. Methods This prospective national multi-centered study used alpha-fetoprotein (AFP) and free β-subunit of human chorionic gonadotrophin( free β-hCG)as the serum markers. From May 2004 to September 2006, 11 centers participated in the collection and analysis of maternal serum AFP and free β-hCG between 14 and 20+6 weeks of pregnancy. The screening results were calculated using the standard algorithm based on the standard database provided with the analytic software. Patients with an increased risk of Down's syndrome pregnancy (≥1/270) were offered genetic anmiocentesis. Outcomes of all pregnancies were obtained.Results A total of 66 132 singleton pregnancies were included in the study. The median maternal age was 27 years. At a cut-eft of 1 in 270, the detection rate (DR) based on a Caucasian database was 72% corresponding to a false positive rate (FPR) of 5%, and the DR based on the Chinese database was raised to 76% corresponding to an FPR of 5%. Conclusion The double-marker test using AFP and free β-hCG is an effective screen strategy for second-trimester detection of fetal Down's syndrome in mainland China. Ethnic variance exists between the Caucasian and Chinese populations. The accuracy of screening is increased by the use of race-specific medians.