Liver histopathological lesions is severe in patients with normal alanine transaminase and low to moderate hepatitis B virus DNA replication

BACKGROUND Chronic hepatitis B virus(HBV)infection remains a major global public health problem.Chronic hepatitis B(CHB)patients can be divided into treatment indication and non-treatment indication individuals according to alanine transaminase(ALT),HBV DNA,serum hepatitis B e antigen status,disease status[liver cirrhosis,hepatocellular carcinoma(HCC),or liver failure],liver necroinflammation or fibrosis,patients’age,and family history of HCC or cirrhosis.For example,normal ALT patients in‘immune-tolerant’phase with HBV DNA higher than 10^(7)or 2×10^(7)IU/mL,and those in‘inactive-carrier’phase with HBV DNA lower than 2×10^(3)IU/mL do not require antiviral therapy.However,is it reasonable to set the defined values of HBV DNA as the fundamental basis to estimate the disease state and to determine whether to start treatment?In fact,we should pay more attention to those who do not match the treatment indications(grayzone patients both in the indeterminate phase and in the‘inactive-carrier’phase).AIM To analyze the correlation of HBV DNA level and liver histopathological severity,and to explore the significance of HBV DNA for CHB with normal ALT.METHODS From January 2017 to December 2021,a retrospective cross-sectional set of 1299 patients with chronic HBV infection(HBV DNA>30 IU/mL)who underwent liver biopsy from four hospitals,including 634 with ALT less than 40 U/L.None of the patients had received anti-HBV treatment.The degrees of liver necroinflammatory activity and liver fibrosis were evaluated according to the Metavir system.On the basis of the HBV DNA level,patients were divided into two groups:Low/moderate replication group,HBV DNA≤10^(7)IU/mL[7.00 Log IU/mL,the European Association for the Study of the Liver(EASL)guidelines]or≤2×10^(7)IU/mL[7.30 Log IU/mL,the Chinese Medical Association(CMA)guidelines];high replication group,HBV DNA>10^(7)IU/mL or>2×10^(7)IU/mL.Relevant factors(demographic characteristics,laboratory parameters and noninvasive models)for liver histopathological severity were analyzed by univariate analysis,logistics analysis and propensity score-matched analysis.RESULTS At entry,there were 21.45%,24.29%,and 30.28%of the patients had liver histopathological severities with≥A2,≥F2,and≥A2 or/and≥F2,respectively.HBV DNA level(negative correlation)and noninvasive model liver fibrosis 5 value(positive correlation)were independent risk factors for liver histopathological severities(liver necroinflammation,liver fibrosis,and treatment indication).The AUROCs of the prediction probabilities(PRE_)of the models mentioned above(<A2 vs≥A2,<F2 vs≥F2,<A2 and<F2 vs≥A2 or/and≥F2)were 0.814(95%CI:0.770-0.859),0.824(95%CI:0.785-0.863),and 0.799(95%CI:0.760-0.838),respectively.HBV DNA level(negative correlation)was still an independent risk factor when diagnostic models were excluded,the P values(<A2 vs≥A2,<F2 vs≥F2,<A2 and<F2 vs≥A2 or/and≥F2)were 0.011,0.000,and 0.000,respectively.For the propensity score-matched pairs,whether based on EASL guidelines or CMA guidelines,the group with significant liver histology damage(≥A2 or/and≥F2)showed much lower HBV DNA level than the group with non-significant liver histology damage(<A2 and<F2).Patients in the moderate replication group(with indeterminate phase)had the most serious liver disease pathologically and hematologically,followed by patients in the low replication group(with‘inactive-carrier’phase)and then the high replication group(with‘immune-tolerant’phase).CONCLUSION HBV DNA level is a negative risk factor for liver disease progression.The phase definition of CHB may be revised by whether the level of HBV DNA exceeds the detection low limit value.Patients who are in the indeterminate phase or‘inactive carriers’should receive antiviral therapy.

Automatic"Ground Truth"Annotation and Industrial Workpiece Dataset Generation for Deep Learning

In industry,it is becoming common to detect and recognize industrial workpieces using deep learning methods.In this field,the lack of datasets is a big problem,and collecting and annotating datasets in this field is very labor intensive.The researchers need to perform dataset annotation if a dataset is generated by themselves.It is also one of the restrictive factors that the current method based on deep learning cannot expand well.At present,there are very few workpiece datasets for industrial fields,and the existing datasets are generated from ideal workpiece computer aided design(CAD)models,for which few actual workpiece images were collected and utilized.We propose an automatic industrial workpiece dataset generation method and an automatic ground truth annotation method.Included in our methods are three algorithms that we proposed:a point cloud based spatial plane segmentation algorithm to segment the workpieces in the real scene and to obtain the annotation information of the workpieces in the images captured in the real scene;a random multiple workpiece generation algorithm to generate abundant composition datasets with random rotation workpiece angles and positions;and a tangent vector based contour tracking and completion algorithm to get improved contour images.With our procedures,annotation information can be obtained using the algorithms proposed in this paper.Upon completion of the annotation process,a json format file is generated.Faster R-CNN(Faster R-convolutional neural network),SSD(single shot multibox detector)and YOLO(you only look once:unified,real-time object detection)are trained using the datasets proposed in this paper.The experimental results show the effectiveness and integrity of this dataset generation and annotation method.