Application and evaluation of artificial intelligence TPS-assisted cytologic screening system in urine exfoliative cytology / 中华病理学杂志

Objective: To explore the application of manual screening collaborated with the Artificial Intelligence TPS-Assisted Cytologic Screening System in urinary exfoliative cytology and its clinical values. Methods: A total of 3 033 urine exfoliated cytology samples were collected at the Henan People's Hospital, Capital Medical University, Beijing, China. Liquid-based thin-layer cytology was prepared. The slides were manually read under the microscope and digitally presented using a scanner. The intelligent identification and analysis were carried out using an artificial intelligence TPS assisted screening system. The Paris Report Classification System of Urinary Exfoliated Cytology 2022 was used as the evaluation standard. Atypical urothelial cells and even higher grade lesions were considered as positive when evaluating the recognition sensitivity, specificity, and diagnostic accuracy of artificial intelligence-assisted screening systems and human-machine collaborative cytologic screening methods in urine exfoliative cytology. Among the collected cases, there were also 1 100 pathological tissue controls. Results: The accuracy, sensitivity and specificity of the AI-assisted cytologic screening system were 77.18%, 90.79% and 69.49%; those of human-machine coordination method were 92.89%, 99.63% and 89.09%, respectively. Compared with the histopathological results, the accuracy, sensitivity and specificity of manual reading were 79.82%, 74.20% and 95.80%, respectively, while those of AI-assisted cytologic screening system were 93.45%, 93.73% and 92.66%, respectively. The accuracy, sensitivity and specificity of human-machine coordination method were 95.36%, 95.21% and 95.80%, respectively. Both cytological and histological controls showed that human-machine coordination review method had higher diagnostic accuracy and sensitivity, and lower false negative rates. Conclusions: The artificial intelligence TPS assisted cytologic screening system has achieved acceptable accuracy in urine exfoliation cytologic screening. The combination of manual screening and artificial intelligence TPS assisted screening system can effectively improve the sensitivity and accuracy of cytologic screening and reduce the risk of misdiagnosis.

Effect of baseline CD(4)(+) T cell count on drop-out of antiretroviral therapy in HIV infected persons in Guangxi Zhuang Autonomous Region, 2008-2015 / 中华流行病学杂志

Objective: To investigate the effect of baseline CD(4)(+) T cell count (CD(4)) on drop-out of antiretroviral therapy (ART) in HIV infected persons. Methods: Retrospective cohort was conducted in this study. HIV infected persons aged≥18 years and receiving free ART for the first time in Guangxi Zhuang Autonomous Region (Guangxi) from 2008 to 2015 were selected from the antiretroviral treatment database of National Comprehensive HIV/AIDS Information System, with follow-up conducted till May 30, 2016. Cause-specific Cox proportional hazard models were used to evaluate effect of different CD(4) on the drop-out of ART in the HIV infected persons. Results: A total of 58 502 eligible study participants were included in this retrospective cohort study. The average drop-out ratio was 4.8/100 person-years. After controlling the following baseline covariates: age, sex, marital status, route of HIV infection, WHO clinical stage before ART, initial/current ART regiment, ART regiment adjustment, and year of initiating ART for potential confounding, the adjusted HR of drop-out for HIV infected persons with 200- cells/μl, 351-cells/μl and ≥500 cells/μl were 1.110 (95%CI: 1.053-1.171, P<0.001), 1.391 (95%CI: 1.278-1.514, P<0.001) and 1.695 (95%CI: 1.497-1.918, P<0.001), respectively, in risk for drop-out compared with those with baseline CD(4)<200 cells/μl. Among the HIV infected persons, 56.0% (1 601/2 861) of drug withdrawal was due to poor compliance with medication. Conclusions: With the increase of baseline CD(4) when initiating ART, the risk for the drop-out in HIV infected persons increased significantly. To further reduce the drop-out of ART, it is important to take CD(4) into account in initiating ART and to strengthen the health education on treatment compliancy and training for healthcare providers.

Impact of HIV and Mycobacterium tuberculosis co-infection on related mortality / 中华流行病学杂志

Objective: To understand the impact of HIV and Mycobacterium tuberculosis (MTB) co-infectious (HIV/MTB) on related mortality in Guangxi Zhuang Autonomous Region, provide evidence for the development of a better HIV/MTB co-infection control and prevention program. Methods: A multiple cross-systems check (MCSC) approach was used to confirm the HIV/MTB co-infection individuals on data related to treatment, follow-up, epidemiological comprehensive and Tuberculosis (TB) special report system. Social demography characteristics, incidence of TB among HIV positive individuals, HIV incidence among MTB infection persons etc., were described. We compared the mortalities and related risks between HIV/MTB co-infection and mono HIV positive individuals as well as between the HIV/MTB co-infection and mono MTB infection persons, using both the Chi Square test and the Cox's proportional hazard regression model (Cox). Results: Reported data showed that the incidence of MTB co-infection in the HIV cohort was 17.72% (2 533/14 293), while HIV incidence in the TB patients was 5.57% (2 351/42 205), respectively. The mortality of HIV/MTB co-infection in the HIV/AIDS cohort was 15.16% (384/2 533) within one-year of observation and was significantly higher than the mortality (13.63%,1 603/11 760) of mono HIV positive individuals (P<0.000 1). The percentage of the HIV/AIDS death cases was 19.33% (384/1 987) who registered and died in the 2011 calendar year were caused by MTB co-infection. Among all the HIV/MTB co-infection patients who had been identified from the HIV cohort, 60.05% (1 521/2 533) had initiated ART, 15.48% (392/2 533) had been cured for TB and 27.48% (696/2 533) had been under complete TB regimen. Among the confirmed HIV/MTB cases from the TB cohort, the cure rate of TB was 19.70% (463/2 351) and the percentage of completed TB regimen was 37.26% (876/2 351). The percentage of the individuals whose CD(4)(+) T lymphocyte cells count appeared less than 200 cell/μl was 64.13% (785/1 224), upon the HIV diagnoses were made. Compared with individuals who were under mono HIV infection, the mortality risk on HIV/MTB co-infection was 1.17 times higher during the five-year observation period, then the patients with only mono MTB infection and the mortality risk in patients with HIV/MTB co-infection was 25.68 times higher under the 12-month observation period. Conclusions: Both the incidence and mortality of HIV/MTB appeared high in Guangxi, with mortality and the risk of mortality in the HIV/MTB co-infection group significantly higher than that in both the HIV mono infection and the MTB mono infections groups. Both the rate of antiretroviral treatment coverage and the cure rate of TB treatment should be increased in no time as well as the capability of early TB case-finding among people living with HIV.

Prevalence and associated risk factors on preterm birth, low birth weight, and small for gestational age among HIV-infected pregnant women in Hunan province, 2011-2017 / 中华流行病学杂志

Objective: To describe the prevalence of preterm birth (PB), low birth weight (LBW), and small for gestational age (SGA) among HIV-infected pregnant women and to identify associated risk factors in Hunan province. Methods: This study appeared a retrospective one on HIV-infected pregnant women retrieved from Information System of Prevention of Mother-to-child Transmission of HIV management in Hunan province, between January 2011 and December 2017. Information regarding demographic characteristics, pregnancy, antiretroviral therapy (ART), husbands/partners' relevant situation and pregnancy outcomes, among these HIV-infected pregnant women were collected and analyzed. The incidence rates on PB, LBW and SGA were calculated. Multivariate logistic regression was used to analyze the associated risk factors. Results: A total of 780 HIV-infected pregnant women were enrolled. The prevalence rates on PB, LBW and SGA in HIV- infected pregnant women appeared as 7.9% (62/780), 9.9% (77/780) and 21.3% (166/780), respectively. Results from the multivariate logistic regression analysis showed that factors as pregnancy related diseases as moderate/severe anemia, hypertensive, initial time of ART <14 gestational weeks (compared to those women without ART during pregnancy) and husbands/partners' age >35 years old (compared to husbands/partners' age 26-30 years old) etc., were associated with an increased risk of PB with adjusted OR as 4.59 (95%CI: 1.51-13.95), 4.90 (95%CI: 1.56-15.46), 2.40 (95%CI: 1.26- 4.56) and 2.29 (95%CI: 1.21-4.36). For LBW, pregnancy moderate/severe anemia, pregnancy HBV infection and initial time of ART <14 gestational weeks were associated with an increased risk of LBW, with adjusted OR as 3.28 (95%CI: 1.13-9.54), 4.37 (95%CI: 1.42-13.44) and 2.68 (95%CI: 1.51-4.76), respectively. For SGA, pregnancy HBV infection and initial time of ART <14 gestational weeks were risk factors for SGA, with adjusted OR as 4.41 (95%CI: 1.43-13.63) and 2.67 (95%CI: 1.51-4.73), respectively. Conclusion: Preterm birth, LBW and SGA were common adverse pregnancy outcomes for HIV-infected pregnant women and were associated with factors as pregnancy complications, ART and husbands/partners' age.

Immunogenicity of inacitivated quadrivalent influenza vaccine in adults aged 18-64 years: A systematic review and Meta-analysis / 中华流行病学杂志

Objective: To evaluate the immunogenicity of inactivated quadrivalent influenza vaccine (QIV) in adults aged 18-64 years, through a Meta-analysis. Methods: Literature was retrieved by searching the Medline, Cochrane Library, Science Direct in the past decade. All the studies were under random control trial (RCT) and including data related to immunogenicity which involving sero-protection rate (SPR) and sero-conversion rate (SCR) of the QIV, versus inactivated trivalent influenza vaccine (TIV) in the population aged 18 to 64. Revman 5.3 software was employed to manipulate the pooled date of the included literature. Result: A total of 8 studies for the SPR and SCR of the shared strains (two A lineage and one B lineage) were included. There appeared no significant differences in the response rates between the two vaccines. As for QIV versus TIV (B/Yamagata), the pooled RR of the SPR for B/Victoria was 1.28 (95%CI: 1.08-1.51, P<0.05), with the pooled RR of the SCR for B/Victoria as 1.94 (95%CI: 1.50-2.50, P<0.05). For QIV versus TIV (B/Victoria), the pooled RR of the SPR for B/Yamagata as 1.10 (95%CI: 1.02-1.18, P<0.05), and the pooled RR of SCR for B/Yamagata as 1.99 (95%CI: 1.34-2.97, P<0.05). Conclusion: In the population aged 18-64 years, inactivated QIV was equivalently immunogenic against the shared three strains included in the activated TIV while a superior immunogenic effect was noticed in the vaccine strain which did not include the inactivated QIV.