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
. 2022 Dec 19:1-29.
doi: 10.1007/s11042-022-14316-7. Online ahead of print.

A new model for classification of medical CT images using CNN: a COVID-19 case study

Pedro Moises de Sousa  1 , Pedro Cunha Carneiro  1 , Gabrielle Macedo Pereira  1 , Mariane Modesto Oliveira  1 , Carlos Alberto da Costa Junior  1 , Luis Vinicius de Moura  2 , Christian Mattjie  2 , Ana Maria Marques da Silva  2 , Túlio Augusto Alves Macedo  3 , Ana Claudia Patrocinio  1
Affiliations

A new model for classification of medical CT images using CNN: a COVID-19 case study

Pedro Moises de Sousa et al. Multimed Tools Appl. .

Abstract

SARS-CoV-2 is the causative agent of COVID-19 and leaves characteristic impressions on chest Computed Tomography (CT) images in infected patients and this analysis is performed by radiologists through visual reading of lung images, and failures may occur. In this article, we propose a classification model, called Wavelet Convolutional Neural Network (WCNN) that aims to improve the differentiation of images of patients with COVID-19 from images of patients with other lung infections. The WCNN model was based on a Convolutional Neural Network (CNN) and wavelet transform. The model proposes a new input layer added to the neural network, which was called Wave layer. The hyperparameters values were defined by ablation tests. WCNN was applied to chest CT images to images from two internal and one external repositories. For all repositories, the average results of Accuracy (ACC), Sensitivity (Sen) and Specificity (Sp) were calculated. Subsequently, the average results of the repositories were consolidated, and the final values were ACC = 0.9819, Sen = 0.9783 and Sp = 0.98. The WCNN model uses a new Wave input layer, which standardizes the network input, without using data augmentation, resizing and segmentation techniques, maintaining the integrity of the tomographic image analysis. Thus, applications developed based on WCNN have the potential to assist radiologists with a second opinion in the analysis.1.

Keywords: COVID-19; Chest CT images; Convolutional neural networks; WCNN; Wavelet.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest and ethical standardsAll authors declare no conflict of interest, and this article does not contain studies with human or animal participants performed by any of the authors.

Figures

Fig. 1
Fig. 1
a Transformation step on each line b Transformation step on each column c Diagram overview of the wavelet transform
Fig. 2
Fig. 2
Workflow overview and the WCNN template
Fig. 3
Fig. 3
Selection of CT slices
Fig. 4
Fig. 4
a Conventional CNN architecture and b Architecture of our customized CNN highlighting the wave layer (Adapted from [20])
Fig. 5
Fig. 5
WCNN classification scheme
Fig. 6
Fig. 6
WCNN Wave Layer
Fig. 7
Fig. 7
WCNN training loss and training accuracy of Dataset I
Fig. 8
Fig. 8
Dataset I confusion matrix
Fig. 9
Fig. 9
Dataset I ROC curve
Fig. 10
Fig. 10
WCNN training loss and training accuracy of Dataset II
Fig. 11
Fig. 11
Dataset II confusion matrix
Fig. 12
Fig. 12
Dataset II ROC curve
Fig. 13
Fig. 13
Average accuracy for internal (Dataset I e Dataset II) and external (Dataset III)
Fig. 14
Fig. 14
Region of Interest (ROI) Selection
Fig. 15
Fig. 15
Histogram sample from original image and corresponding diagonal, vertical, and approximation coefficients
See this image and copyright information in PMC

References

    1. Aggarwal CC, et al. Neural networks and deep learning. Springer; 2018.
    1. Balas VE, Roy SS, Sharma D, Samui P. Handbook of deep learning applications. Springer; 2019.
    1. Chen J, Wu L, Zhang J, Zhang L, Gong D, Zhao Y, Chen Q, Huang S, Yang M, Yang X, Hu S, Wang Y, Hu X, Zheng B, Zhang K, Wu H, Dong Z, Xu Y, Zhu Y, Chen X, Zhang M, Yu L, Cheng F, Yu H. Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography. Sci Rep. 2020;10:1–11. doi: 10.1038/s41598-020-76282-0. - DOI - PMC - PubMed
    1. Chen YM, Chen YJ, Ho WH, Tsai JT. Classifying chest CT images as COVID-19 positive/negative using a convolutional neural network ensemble model and uniform experimental design method. BMC Bioinformatics. 2021;22:1–19. doi: 10.1186/s12859-021-04083-x. - DOI - PMC - PubMed
    1. Dai W, Zhang H, Yu J, Xu H, Chen H, Luo S, Zhang H, Liang L, Wu X, Lei Y, Lin F. CT imaging and differential diagnosis of COVID-19. Can Assoc Radiol J. 2020;71:195–200. doi: 10.1177/0846537120913033. - DOI - PMC - PubMed

LinkOut - more resources