2dCNN-BiCuDNNLSTM: Hybrid Deep-Learning-Based Approach for Classification of COVID-19 X-ray Images
Abstract
:1. Introduction
- By analyzing the association of image data, a new deep learning method, 2dCNN-BiCuDNNLSTM, is proposed for COVID-19 and viral pneumonia classification.
- By comparing with the other deep learning methods, such as stacked 2dCNN, for COVID-19 and viral pneumonia detection and classification, it is established that the 2dCNN-BiCuDNNLSTM network is the most precise and effective, which shows that it is more suitable for COVID-19 and viral pneumonia classification.
2. Literature Review
3. Methodology
3.1. BiCuDNNLSTM (Bidirectional Cuda Based Long Short Term Memory)
3.2. 2D CNN (2-Dimensional Convolution Neural Network)
3.3. 2dCNN-BiCuDNNLSTM Model Architecture
Algorithm 1 Hybrid 2dCNN-BiCuDNNLSTM Classification Model. |
Input: Image Dataset of Covid Cases, Viral Pneumonia and Healthy Cases Image Augmentation: ImageDataGenerator while do while batch-size = 1 : b do Sequential() Conv2D MaxPool2D Dropout(0.2) Conv2D MaxPool2D Dropout(0.2) Conv2D MaxPool2D Dropout(0.2) Conv2D MaxPool2D Reshape Bidirectional(CuDNNLSTM Dropout(0.2) Dense Dense Compile end while end while |
- Kernel size: The Kernel size refers to the width and height of the filter cover.
- Activation: This function is used to decide which neuron should be activated or not, by evaluating the weighted sum and, further, adding bias with it.
- Strides: Stride is the number of pixel transfers over the input matrix.
- Optimizer: An optimizer is a function used to reshape the features of the model, such as weights and learning rate. It supports in decreasing loss and improving model accuracy.
- Pool size: A pooling function is used to choose the maximum segment from the region of the attribute map masked by the filter.
- Loss: The loss function is used to calculate the difference between the current output of the model and the expected output.
- Metrics: Metrics are used to evaluate the performance of proposed model.
- Batch size: Batch size defines the number of training examples that will be trained through the model in one iteration.
- Epoch: The number of epochs are the number of entire passes through the training dataset, for one cycle.
3.4. Evaluation Metric
- Accuracy: Acuracy is used to judge the potential of a network, by calculating a proportion of the accurately predicted cases from the total number of cases. Accuracy is demonstrated as:
- True Positive (TP): Portion of correctly predicted +ive cases
- True Negative (TN): Portion of correctly predicted -ive cases
- False Positive (FP): Portion of incorrectly predicted +ive cases
- True Negative (TN): Portion of incorrectly predicted -ive cases
- Precision: Precision is the proportion of accurately predicted +ive observations, from the total predicted +ive cases. It is demonstrated as:
- Recall: Recall is the proportion of accurately predicted +ive cases, from all cases in the original class.
- F1 score: F1 score contributes to the balance among Precision and Recall.
4. Experiments
4.1. Dataset
4.2. Dataset Augmentation
4.3. Training and Classification Processes
4.4. Model’s Parameters
5. Results and Discussion
Comparison with Other CNN and BiCuDNNLSTM Architectures
6. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
Abbreviations
DL | Deep Learning |
2dCNN | Two-Dimensional Convolutional Neural Network |
BiCuDNNLSTM | Bidirectional CUDA Deep Neural Network Long Short Term Memory |
RMSE | Root Mean Square Error |
TP | True Positive |
TN | True Negative |
FP | False Positive |
TN | True Negative |
GPUs | Graphics Processor Unit |
TPUs | Tensor Processing Units |
References
- Kesim, E.; Dokur, Z.; Olmez, T. X-ray Chest Image Classification by A Small-Sized Convolutional Neural Network. In Proceedings of the 2019 Scientific Meeting on Electrical-Electronics & Biomedical Engineering and Computer Science (EBBT), Istanbul, Turkey, 24–26 April 2019. [Google Scholar] [CrossRef]
- Zou, L.; Zheng, J.; Miao, C.; Mckeown, M.J.; Wang, Z.J. 3D CNN Based Automatic Diagnosis of Attention Deficit Hyperactivity Disorder Using Functional and Structural MRI. IEEE Access 2017, 5, 23626–23636. [Google Scholar] [CrossRef]
- Liu, C.; Cao, Y.; Alcantara, M.; Liu, B.; Brunette, M.; Peinado, J. TX-CNN: Detecting tuberculosis in chest X-ray images using convolutional neural network. In Proceedings of the 2017 IEEE International Conference on Image Processing (ICIP), Beijing, China, 17–20 September 2017. [Google Scholar] [CrossRef]
- Zhao, X.; Liu, L.; Qi, S.; Teng, Y.; Li, J.; Qian, W. Agile convolutional neural network for pulmonary nodule classification using CT images. Int. J. Comput. Assist. Radiol. Surg. 2018, 13, 585–595. [Google Scholar] [CrossRef] [PubMed]
- Chen, J.; Wang, Y.; Wu, Y.; Cai, C. An Ensemble of Convolutional Neural Networks for Image Classification Based on LSTM. In Proceedings of the 2017 International Conference on Green Informatics (ICGI), Fuzhou, China, 15–17 August 2017. [Google Scholar] [CrossRef]
- Li, P.; Tang, H.; Yu, J.; Song, W. LSTM and multiple CNNs based event image classification. Multimed. Tools Appl. 2021, 80, 30743–30760. [Google Scholar] [CrossRef]
- Chen, Y.; Jiang, H.; Li, C.; Jia, X.; Ghamisi, P. Deep Feature Extraction and Classification of Hyperspectral Images Based on Convolutional Neural Networks. IEEE Trans. Geosci. Remote Sens. 2016, 54, 6232–6251. [Google Scholar] [CrossRef] [Green Version]
- Apostolopoulos, I.D.; Bessiana, T.A. Covid-19: Automatic detection from X-ray images utilizing Transfer Learning with Convolutional Neural Networks. Phys. Eng. Sci. Med. 2020, 43, 635–640. [Google Scholar] [CrossRef] [Green Version]
- Wang, L.; Lin, Z.Q.; Wong, A. COVID-Net: A tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images. Sci. Rep. 2020, 10, 19549. [Google Scholar] [CrossRef]
- Afshar, P.; Heidarian, S.; Naderkhani, F.; Oikonomou, A.; Plataniotis, K.N.; Mohammadi, A. COVID-CAPS: A capsule network-based framework for identification of COVID-19 cases from X-ray images. Pattern Recognit. Lett. 2020, 138, 638–643. [Google Scholar] [CrossRef]
- Apostolopoulos, I.D.; Aznaouridis, S.I.; Tzani, M. Extracting Possibly Representative COVID-19 Biomarkers from X-ray Images with Deep Learning Approach and Image Data Related to Pulmonary Diseases. J. Med. Biol. Eng. 2020, 40, 462–469. [Google Scholar] [CrossRef]
- Zhao, W.; Jiang, W.; Qiu, X. Deep learning for COVID-19 detection based on CT images. Sci. Rep. 2021, 11, 14353. [Google Scholar] [CrossRef]
- Al-Waisy, A.S.; Mohammed, M.A.; Mohammed, M.A.; Al-Fahdawi, S.; Al-Fahdawi, S.; Maashi, M.; Zapirain, B.G.; Abdulkareem, K.H.; Mostafa, S.; Kumar, N.M.; et al. COVID-DeepNet Hybrid Multimodal Deep Learning System for Improving COVID-19 Pneumonia Detection in Chest X-ray Images. Comput. Mater. Contin. 2021, 67, 2409–2429. [Google Scholar] [CrossRef]
- Khuzani, A.Z.; Heidari, M.; Shariati, S.A. A comprehensive investigation of machine learning feature extraction and classification methods for automated diagnosis of COVID-19 based on X-ray images. Sci. Rep. 2021, 11, 9887. [Google Scholar] [CrossRef] [PubMed]
- Hu, S.; Gao, Y.; Niu, Z.; Jiang, Y.; Li, L.; Xiao, X.; Wang, M.; Fang, E.F.; Menpes-Smith, W.; Xia, J.; et al. Weakly Supervised Deep Learning for COVID-19 Infection Detection and Classification From CT Images. IEEE Access 2020, 8, 118869–118883. [Google Scholar] [CrossRef]
- Ozturk, T.; Talo, M.; Yildirim, E.A.; Baloglu, U.B.; Yildirim, O.; Acharya, U. Automated detection of COVID-19 cases using deep neural networks with X-ray images. Comput. Biol. Med. 2020, 121, 103792. [Google Scholar] [CrossRef]
- Muthumayil, K.; Buvana, M.; Sekar, K.R.; Amraoui, A.E.; Nouaouri, I.; Mansour, R.F. Optimized convolutional neural network for automatic detection of COVID-19. Comput. Mater. Contin. 2021, 70, 1159–1175. [Google Scholar] [CrossRef]
- Xue, Y.; Onzo, B.M.; Mansour, R.F.; Su, S. Deep Convolutional Neural Network Approach for COVID-19 Detection. Comput. Syst. Sci. Eng. 2022, 42, 201–211. [Google Scholar] [CrossRef]
- Ismael, A.M.; Kadir Şengür, A. Deep learning approaches for COVID-19 detection based on chest X-ray images. Expert Syst. Appl. 2021, 164, 114054. [Google Scholar] [CrossRef] [PubMed]
- Oh, Y.; Park, S.; Ye, J.C. Deep Learning COVID-19 Features on CXR Using Limited Training Data Sets. IEEE Trans. Med. Imaging 2020, 39, 2688–2700. [Google Scholar] [CrossRef]
- Biswas, S.; Dash, S. LSTM-CNN Deep Learning–Based Hybrid System for Real-Time COVID-19 Data Analysis and Prediction Using Twitter Data; Springer: Berlin/Heidelberg, Germany, 2021. [Google Scholar] [CrossRef]
- Jain, G.; Mittal, D.; Thakur, D.; Mittal, M.K. A deep learning approach to detect Covid-19 coronavirus with X-ray images. Biocybern. Biomed. Eng. 2020, 40, 1391–1405. [Google Scholar] [CrossRef]
- Hochreiter, S.; Schmidhuber, J. Long Short-Term Memory. Neural Comput. 1997, 9, 1735–1780. [Google Scholar] [CrossRef]
- Yadav, A.; Jhaa, C.K.; Sharanb, A. Optimizing LSTM for time series prediction in Indian stock market. Procedia Comput. Sci. 2019, 167, 2091–2100. [Google Scholar] [CrossRef]
- Aveleira-Mata, J.; Ondicol-Garcia, J.; Munoz-Castaneda, A.L.; Garcia, I.; Benavides, C. Multiclass Classification Procedure for Detecting Attacks on MQTT-IoT Protocol. Adv. Complex Syst. Their Appl. Cybersecur. 2019, 2019, 6516253. [Google Scholar] [CrossRef] [Green Version]
- Appleyard, J.; Kocisky, T.; Blunsom, P. Optimizing Performance of Recurrent Neural Networks on GPUs. 2016. Available online: http://xxx.lanl.gov/abs/1604.01946 (accessed on 31 December 2021).
- Sezer, O.B.; Gudelek, M.U.; Ozbayoglu, A.M. Financial time series forecasting with deep learning: A systematic literature review: 2005–2019. Appl. Soft Comput. 2020, 90, 106181. [Google Scholar] [CrossRef] [Green Version]
- Silva, P.; Luz, E.; Silva, G.; Moreira, G.; Silva, R.; Lucio, D.; Menotti, D. COVID-19 detection in CT images with deep learning: A voting-based scheme and cross-datasets analysis. Biocybern. Biomed. Eng. 2020, 20, 100427. [Google Scholar] [CrossRef] [PubMed]
Precision | Recall | F1 Score | |
---|---|---|---|
COVID-19 | 100 | 100 | 100 |
Normal | 94 | 80 | 86 |
Viral Pneumonia | 83 | 95 | 88 |
Accuracy | 93 | ||
Macro Average | 92 | 92 | 92 |
Weighted Average | 93 | 92 | 92 |
Approach | Training Accuracy | Validation Accuracy | Test Accuracy |
---|---|---|---|
2dCNN-BiCuDNNLSTM | 87 | 86 | 93 |
Approach | Accuracy | Precision | Recall | F1 Score |
---|---|---|---|---|
2dCNN-BiCuDNNLSTM | 93 | 100 | 100 | 100 |
2dCNN | 38 | 58 | 96 | 72 |
Approach | Accuracy | Precision | Recall | F1 Score |
---|---|---|---|---|
2dCNN-BiCuDNNLSTM | 93 | 83 | 95 | 88 |
2dCNN | 38 | 37 | 93 | 53 |
Approach | Accuracy | Precision | Recall | F1 Score |
---|---|---|---|---|
2dCNN-BiCuDNNLSTM | 93 | 94 | 80 | 86 |
2dCNN | 38 | 37 | 93 | 53 |
Approach | Accuracy |
---|---|
2dCNN-BiCuDNNLSTM | 88.06 |
DarkCovidNet model [16] | 87.02 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Kanwal, A.; Chandrasekaran, S. 2dCNN-BiCuDNNLSTM: Hybrid Deep-Learning-Based Approach for Classification of COVID-19 X-ray Images. Sustainability 2022, 14, 6785. https://doi.org/10.3390/su14116785
Kanwal A, Chandrasekaran S. 2dCNN-BiCuDNNLSTM: Hybrid Deep-Learning-Based Approach for Classification of COVID-19 X-ray Images. Sustainability. 2022; 14(11):6785. https://doi.org/10.3390/su14116785
Chicago/Turabian StyleKanwal, Anika, and Siva Chandrasekaran. 2022. "2dCNN-BiCuDNNLSTM: Hybrid Deep-Learning-Based Approach for Classification of COVID-19 X-ray Images" Sustainability 14, no. 11: 6785. https://doi.org/10.3390/su14116785