Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Jun 10;9(6):e99422.
doi: 10.1371/journal.pone.0099422. eCollection 2014.

Artificial neural network accurately predicts hepatitis B surface antigen seroclearance

Ming-Hua Zheng  1 , Wai-Kay Seto  2 , Ke-Qing Shi  3 , Danny Ka-Ho Wong  2 , James Fung  2 , Ivan Fan-Ngai Hung  4 , Daniel Yee-Tak Fong  5 , John Chi-Hang Yuen  4 , Teresa Tong  4 , Ching-Lung Lai  2 , Man-Fung Yuen  2
Affiliations
Free PMC article

Artificial neural network accurately predicts hepatitis B surface antigen seroclearance

Ming-Hua Zheng et al. PLoS One. .
Free PMC article

Abstract

Background & aims: Hepatitis B surface antigen (HBsAg) seroclearance and seroconversion are regarded as favorable outcomes of chronic hepatitis B (CHB). This study aimed to develop artificial neural networks (ANNs) that could accurately predict HBsAg seroclearance or seroconversion on the basis of available serum variables.

Methods: Data from 203 untreated, HBeAg-negative CHB patients with spontaneous HBsAg seroclearance (63 with HBsAg seroconversion), and 203 age- and sex-matched HBeAg-negative controls were analyzed. ANNs and logistic regression models (LRMs) were built and tested according to HBsAg seroclearance and seroconversion. Predictive accuracy was assessed with area under the receiver operating characteristic curve (AUROC).

Results: Serum quantitative HBsAg (qHBsAg) and HBV DNA levels, qHBsAg and HBV DNA reduction were related to HBsAg seroclearance (P<0.001) and were used for ANN/LRM-HBsAg seroclearance building, whereas, qHBsAg reduction was not associated with ANN-HBsAg seroconversion (P = 0.197) and LRM-HBsAg seroconversion was solely based on qHBsAg (P = 0.01). For HBsAg seroclearance, AUROCs of ANN were 0.96, 0.93 and 0.95 for the training, testing and genotype B subgroups respectively. They were significantly higher than those of LRM, qHBsAg and HBV DNA (all P<0.05). Although the performance of ANN-HBsAg seroconversion (AUROC 0.757) was inferior to that for HBsAg seroclearance, it tended to be better than those of LRM, qHBsAg and HBV DNA.

Conclusions: ANN identifies spontaneous HBsAg seroclearance in HBeAg-negative CHB patients with better accuracy, on the basis of easily available serum data. More useful predictors for HBsAg seroconversion are still needed to be explored in the future.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have the following interests: MF Yuen and J Fung are academic editors of PLOS ONE. The assays used to determine serum hepatitis B virus DNA level (Cobas Taqman assay) and hepatitis B surface antigen level (Elecsys HBsAg II assay) performed in our laboratory were supported by an unrestricted grant from Roche Diagnostics. There are no patents, products in development or marketed products to declare. This does not alter the authors' adherence to all PLOS ONE policies on sharing data and materials.

Figures

Figure 1
Figure 1. Schematic representation of the artificial neural network developed to predict A) HBsAg seroclearance and B) HBsAg seroconversion.
Note: qHBsAg, 3 years before baseline; HBV DNA, 3 years before baseline; qHBsAg reduction, 3 to 2 years before baseline; HBV DNA reduction, 3 to 2 years before baseline.
Figure 2
Figure 2. Relative weights of the clinical input parameters in an artificial neural network (ANN) trained with patients of training set.
(A) ANN for HBsAg seroclearance; (B) ANN for HBsAg seroconversion. Note: qHBsAg, 3 years before baseline; HBV DNA, 3 years before baseline; qHBsAg reduction, 3 to 2 years before baseline; HBV DNA reduction, 3 to 2 years before baseline.
Figure 3
Figure 3. ROC analysis displaying the ability of four models/parameters (ANN, LRM, qHBsAg* and HBV DNA*) to discriminate HBsAg seroclearance in (A) training group; (B) testing group; (C) genotype B subgroup; (D) genotype C subgroup.
*Time point 3 years; LRM, logistic regression model; ANN, artificial neural network.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Lok AS, McMahon BJ (2007) Chronic hepatitis B. Hepatology 45: 507–539. - PubMed
    1. Tseng TC, Kao JH (2009) HBsAg seroclearance: the more and earlier, the better. Gastroenterology 136: : 1842–1843; author reply 1843–1844. - PubMed
    1. Kobayashi M, Hosaka T, Suzuki F, Akuta N, Sezaki H, et al. (2014) Seroclearance rate of hepatitis B surface antigen in 2,112 patients with chronic hepatitis in Japan during long-term follow-up. J Gastroenterol 49: 538–546. - PubMed
    1. Arai M, Togo S, Kanda T, Fujiwara K, Imazeki F, et al. (2012) Quantification of hepatitis B surface antigen can help predict spontaneous hepatitis B surface antigen seroclearance. Eur J Gastroenterol Hepatol 24: 414–418. - PubMed
    1. Tseng TC, Liu CJ, Su TH, Wang CC, Chen CL, et al.. (2011) Serum hepatitis B surface antigen levels predict surface antigen loss in hepatitis B e antigen seroconverters. Gastroenterology 141: : 517–525, 525 e511–512. - PubMed

Publication types

Substances

Grants and funding

The assays used to determine serum hepatitis B virus DNA level (Cobas Taqman assay) and hepatitis B surface antigen level (Elecsys HBsAg II assay) performed in our laboratory were supported by an unrestricted grant from Roche Diagnostics. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

LinkOut - more resources