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
. 2021 Feb 26;22(18):17431-17438.
doi: 10.1109/JSEN.2021.3062442. eCollection 2022 Sep.

AVNC: Attention-Based VGG-Style Network for COVID-19 Diagnosis by CBAM

Shui-Hua Wang  1 , Steven Lawrence Fernandes  2 , Ziquan Zhu  3 , Yu-Dong Zhang  4
Affiliations

AVNC: Attention-Based VGG-Style Network for COVID-19 Diagnosis by CBAM

Shui-Hua Wang et al. IEEE Sens J. .

Abstract

(Aim) To detect COVID-19 patients more accurately and more precisely, we proposed a novel artificial intelligence model. (Methods) We used previously proposed chest CT dataset containing four categories: COVID-19, community-acquired pneumonia, secondary pulmonary tuberculosis, and healthy subjects. First, we proposed a novel VGG-style base network (VSBN) as backbone network. Second, convolutional block attention module (CBAM) was introduced as attention module into our VSBN. Third, an improved multiple-way data augmentation method was used to resist overfitting of our AI model. In all, our model was dubbed as a 12-layer attention-based VGG-style network for COVID-19 (AVNC) (Results) This proposed AVNC achieved the sensitivity/precision/F1 per class all above 95%. Particularly, AVNC yielded a micro-averaged F1 score of 96.87%, which is higher than 11 state-of-the-art approaches. (Conclusion) This proposed AVNC is effective in recognizing COVID-19 diseases.

Keywords: Attention; VGG; convolutional block attention module; convolutional neural network; covid-19; diagnosis.

PubMed Disclaimer

Figures

Fig. 1.
Fig. 1.
Preprocessing on raw dataset (HS: histogram stretch).
Fig. 2.
Fig. 2.
AM comparison.
Fig. 3.
Fig. 3.
Diagram of an HVS system.
Fig. 4.
Fig. 4.
Relation of CBAM and CAM and SAM.
Fig. 5.
Fig. 5.
Flowchart of two blocks in SAM.
Fig. 6.
Fig. 6.
Illustration of AACB: Integration of CBAM with VGG-style base network.
Fig. 7.
Fig. 7.
AMs of AVNC.
Fig. 8.
Fig. 8.
Diagram of proposed IMDA method.
Fig. 9.
Fig. 9.
formula image result of IMDA.
Fig. 10.
Fig. 10.
Confusion matrix of proposed AVNC approach.
Fig. 11.
Fig. 11.
Comparison (kS: sensitivity of class k, kP: precision of class k, kF: F1 score of class k, mF: micro-averaged F1 score).
Fig. 12.
Fig. 12.
Heatmap generated by Grad-CAM and our AVNC model (HM: heatmap).
See this image and copyright information in PMC

References

    1. Hadi A. G., Kadhom M., Hairunisa N., Yousif E., and Mohammed S. A., “A review on COVID-19: Origin, spread, symptoms, treatment, and prevention,” Biointerface Res. Appl. Chem., vol. 10, pp. 7234–7242, Dec. 2020.
    1. Campos G. S.et al. , “Ion torrent-based nasopharyngeal swab metatranscriptomics in COVID-19,” J. Virolog. Methods, vol. 282, Aug. 2020, Art. no. 113888. - PMC - PubMed
    1. Brammer C.et al. , “Qualitative review of early experiences of off-site COVID-19 testing centers and associated considerations,” Healthcare-J. Del. Sci. Innov., vol. 8, no. 3, Sep. 2020, Art. no. 100449. - PMC - PubMed
    1. Li Y. and Xia L., “Coronavirus disease 2019 (COVID-19): Role of chest CT in diagnosis and management,” Amer. J. Roentgenol., vol. 214, no. 6, pp. 1280–1286, Jun. 2020. - PubMed
    1. Lu S.et al. , “A pathological brain detection system based on extreme learning machine optimized by bat algorithm,” CNS Neurolog. Disorders-Drug Targets, vol. 16, no. 1, pp. 23–29, Jan. 2017. - PubMed

LinkOut - more resources