Application of a light-weighted convolutional neural network for automatic recognition of coal workers' pneumoconiosis in the early stage / 中华劳动卫生职业病杂志

Objective: To construct and verify a light-weighted convolutional neural network (CNN), and explore its application value for screening the early stage (subcategory 0/1 and stage Ⅰ of pneumoconiosis) of coal workers' pneumoconiosis (CWP) from digital chest radiography (DR) . Methods: A total of 1225 DR images of coal workers who were examined at an Occupational Disease Prevention and Control Institute in Anhui Province from October 2018 to March 2021 were retrospectively collected. All DR images were collectively diagnosed by 3 radiologists with diagnostic qualifications and gave diagnostic results. There were 692 DR images with small opacity profusion 0/- or 0/0 and 533 DR images with small opacity profusion 0/1 to stage Ⅲ of pneumoconiosis. The original chest radiographs were preprocessed differently to generate four datasets, namely 16-bit grayscale original image set (Origin16), 8-bit grayscale original image set (Origin 8), 16-bit grayscale histogram equalized image set (HE16) and 8-bit grayscale histogram equalized image set (HE8). The light-weighted CNN, ShuffleNet, was applied to train the generated prediction model on the four datasets separately. The performance of the four models for pneumoconiosis prediction was evaluated on a test set containing 130 DR images using measures such as the receiver operating characteristic (ROC) curve, accuracy, sensitivity, specificity, and Youden index. The Kappa consistency test was used to compare the agreement between the model predictions and the physician diagnosed pneumoconiosis results. Results: Origin16 model achieved the highest ROC area under the curve (AUC=0.958), accuracy (92.3%), specificity (92.9%), and Youden index (0.8452) for predicting pneumoconiosis, with a sensitivity of 91.7%. And the highest consistency between identification and physician diagnosis was observed for Origin16 model (Kappa value was 0.845, 95%CI: 0.753-0.937, P<0.001). HE16 model had the highest sensitivity (98.3%) . Conclusion: The light-weighted CNN ShuffleNet model can efficiently identify the early stages of CWP, and its application in the early screening of CWP can effectively improve physicians' work efficiency.

The median survival and the effect of antiretroviral therapy among HIV-positive former blood donors in Anhui province / 中华流行病学杂志

Objective To estimate the median survival and the effect of antiretroviral therapy (ART) among HIV-positive former blood donors of Fuyang in Anhui province, China. Methods A retrospective survey was conducted among HIV-positive former blood donors, and data was collected on survivors who had received ART. Weibull function was used to calculate median survival of HIV-positive former blood donors. The effect of ART was estimated through comparing the actual number of deaths after ART with the expected number of deaths in those who did not receive ART. Results The median period of HIV infection was at the end of 1994, with the median survival of HIV-positive former blood donors in Fuyang as 10.8 years. By the end of September 2008, among 159 former blood donors, 74 received ART, with their mean CD4+ T-cell count increased from 247.8/μl to 475.1/μl (P<0.0001). 76 of the 159 former blood donors died. When comparing the expected number of deaths calculated by Weibull function, 31.7% of the total number of deaths was reduced. Conclusion Result from this study was in consistent to the UNAIDS' figures that in the absence of treatment, the net median survival time after infection with HIV was estimated to be 11 years and ART has reduced about one third of the expected deaths.

Effect of GBV-C/HIV coinfection on HIV/AIDS disease progression and HIV replication / 病毒学报

Several research groups have recently reported that persistent GB virus C (GBV-C) co-infected with human immunodeficiency virus (HIV) leads to slower AIDSs disease progression than HIV-1 infection alone. However, these findings were not confirmed by several other studies. To investigate the association between GBV-C replication and plasma HIV loads and CD4+ T cell counts, 203 HIV-1 positive former blood/plasma donors(FBDs) were enrolled from Fuyang city of Anhui Province in China. Plasma specimens were collected from them and were tested for GBV-C using RT-PCR and ELISA. Out of 203 specimens, 52 (25.6%) cases were positive for GBV-C, including 35 male (67.3%) and 17 female (32.7%) cases. No significant association was identified between GBV-C infection and CD4+ T-cell counts or between GBV-C infection and HIV viral loads. Since all the subjects studied were naive to ART, the influence of therapy on AIDS disease progression was ruled out in this study. Overall, our data indicated that HIV-1 positive male FBDs were prone to be infected, GBV-C coinfection with HIV-1 does not significantly influence HIV/AIDS disease progression during the late stage of chronic HIV-1 infection.

Epidemiology, clinical and laboratory characteristics of currently alive HIV-1 infected former blood donors naive to antiretroviral therapy in Anhui Province, China / 中华医学杂志(英文版)

<p><b>BACKGROUND</b>Unregulated commercial blood/plasma collection among farmers occurred between 1992 and 1995 in central China and caused the second major epidemic of human immunodeficiency virus type 1 (HIV-1) infection in China. It is important to characterize HIV-1-infected former blood donors and to study characteristics associated with disease progression for future clinical intervention and vaccine development.</p><p><b>METHODS</b>A cross-sectional study was performed on HIV-1-infected former blood donors (FBDs) and age-matched HIV-seronegative local residents. Demographic, epidemiologic, clinical and key laboratory data were collected from all study participants. Both unadjusted and adjusted multivariate linear regressions were employed to analyze the association of the decrease of CD4(+) T-cell counts with other characteristics.</p><p><b>RESULTS</b>Two hundred and ninety-four HIV-1-infected FBDs and 59 age-matched HIV-seronegative local residents were enrolled in this study. The unregulated blood/plasma collection occurred more than a decade (10.8 - 12.8 years) ago, which caused the rapid spread of HIV-1 infection and the high prevalence of co-infection with hepatitis C virus (HCV, 89.5%); hepatitis B virus (HBV) co-infection was observed in only 11 HIV(+)participants (3.7%). Deterioration in both clinical manifestation and laboratory parameters and increase of viral loads were observed in parallel with the decrease of CD4(+) T-cell counts. The decrease of total lymphocyte counts (P < 0.001) and hemoglobin levels (P < 0.001) and the appearance of dermatosis (P = 0.03) were observed in parallel with the decrease of CD4(+) T-cell counts whereas viral loads (P < 0.001) and CD8(+) T-cell counts (P = 0.01) were inversely associated with CD4(+) T-cell counts.</p><p><b>CONCLUSIONS</b>Co-infection with HCV but not HBV is highly prevalent among HIV-1-infected FBDs. CD4(+) T-cell counts is a reliable indicator for disease progression among FBDs. Total lymphocyte counts, hemoglobin level and appearance of dermatosis were positively associated with CD8(+) T-cell counts and viral loads were inversely associated with the decreased CD4(+) T-cell counts.</p>