The Detection of COVID-19 in Chest X-rays Using Ensemble CNN Techniques
Advances in the field of image classification using convolutional neural networks (CNNs) have greatly improved the accuracy of medical image diagnosis by radiologists. Numerous research groups have applied CNN methods to diagnose respiratory illnesses from chest X-rays and have extended this work to...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2078-2489/14/7/370 |
| _version_ | 1797588900431003648 |
|---|---|
| author | Domantas Kuzinkovas Sandhya Clement |
| author_facet | Domantas Kuzinkovas Sandhya Clement |
| author_sort | Domantas Kuzinkovas |
| collection | DOAJ |
| description | Advances in the field of image classification using convolutional neural networks (CNNs) have greatly improved the accuracy of medical image diagnosis by radiologists. Numerous research groups have applied CNN methods to diagnose respiratory illnesses from chest X-rays and have extended this work to prove the feasibility of rapidly diagnosing COVID-19 with high degrees of accuracy. One issue in previous research has been the use of datasets containing only a few hundred images of chest X-rays containing COVID-19, causing CNNs to overfit the image data. This leads to lower accuracy when the model attempts to classify new images, as would be clinically expected. In this work, we present a model trained on the COVID-QU-Ex dataset containing 33,920 chest X-ray images, with an equal share of COVID-19, Non-COVID pneumonia, and Normal images. The model is an ensemble of pre-trained CNNs (ResNet50, VGG19, and VGG16) and GLCM textural features. The model achieved a 98.34% binary classification accuracy (COVID-19/no COVID-19) on a test dataset of 6581 chest X-rays and 94.68% for distinguishing between COVID-19, Non-COVID pneumonia, and normal chest X-rays. The results also demonstrate that a higher 98.82% three-class test accuracy can be achieved using the model if the training dataset only contains a few thousand images. However, the generalizability of the model suffers due to the smaller dataset size. This study highlights the benefits of both ensemble CNN techniques and larger dataset sizes for medical image classification performance. |
| first_indexed | 2024-03-11T00:59:34Z |
| format | Article |
| id | doaj.art-4cda17185d074be09be21d44507a2c51 |
| institution | Directory Open Access Journal |
| issn | 2078-2489 |
| language | English |
| last_indexed | 2024-03-11T00:59:34Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Information |
| spelling | doaj.art-4cda17185d074be09be21d44507a2c512023-11-18T19:46:39ZengMDPI AGInformation2078-24892023-06-0114737010.3390/info14070370The Detection of COVID-19 in Chest X-rays Using Ensemble CNN TechniquesDomantas Kuzinkovas0Sandhya Clement1School of Biomedical Engineering, The University of Sydney, Sydney, NSW 2006, AustraliaSchool of Biomedical Engineering, The University of Sydney, Sydney, NSW 2006, AustraliaAdvances in the field of image classification using convolutional neural networks (CNNs) have greatly improved the accuracy of medical image diagnosis by radiologists. Numerous research groups have applied CNN methods to diagnose respiratory illnesses from chest X-rays and have extended this work to prove the feasibility of rapidly diagnosing COVID-19 with high degrees of accuracy. One issue in previous research has been the use of datasets containing only a few hundred images of chest X-rays containing COVID-19, causing CNNs to overfit the image data. This leads to lower accuracy when the model attempts to classify new images, as would be clinically expected. In this work, we present a model trained on the COVID-QU-Ex dataset containing 33,920 chest X-ray images, with an equal share of COVID-19, Non-COVID pneumonia, and Normal images. The model is an ensemble of pre-trained CNNs (ResNet50, VGG19, and VGG16) and GLCM textural features. The model achieved a 98.34% binary classification accuracy (COVID-19/no COVID-19) on a test dataset of 6581 chest X-rays and 94.68% for distinguishing between COVID-19, Non-COVID pneumonia, and normal chest X-rays. The results also demonstrate that a higher 98.82% three-class test accuracy can be achieved using the model if the training dataset only contains a few thousand images. However, the generalizability of the model suffers due to the smaller dataset size. This study highlights the benefits of both ensemble CNN techniques and larger dataset sizes for medical image classification performance.https://www.mdpi.com/2078-2489/14/7/370automatic diagnosisCOVID-19neural networkartificial intelligenceensemble machine learning |
| spellingShingle | Domantas Kuzinkovas Sandhya Clement The Detection of COVID-19 in Chest X-rays Using Ensemble CNN Techniques Information automatic diagnosis COVID-19 neural network artificial intelligence ensemble machine learning |
| title | The Detection of COVID-19 in Chest X-rays Using Ensemble CNN Techniques |
| title_full | The Detection of COVID-19 in Chest X-rays Using Ensemble CNN Techniques |
| title_fullStr | The Detection of COVID-19 in Chest X-rays Using Ensemble CNN Techniques |
| title_full_unstemmed | The Detection of COVID-19 in Chest X-rays Using Ensemble CNN Techniques |
| title_short | The Detection of COVID-19 in Chest X-rays Using Ensemble CNN Techniques |
| title_sort | detection of covid 19 in chest x rays using ensemble cnn techniques |
| topic | automatic diagnosis COVID-19 neural network artificial intelligence ensemble machine learning |
| url | https://www.mdpi.com/2078-2489/14/7/370 |
| work_keys_str_mv | AT domantaskuzinkovas thedetectionofcovid19inchestxraysusingensemblecnntechniques AT sandhyaclement thedetectionofcovid19inchestxraysusingensemblecnntechniques AT domantaskuzinkovas detectionofcovid19inchestxraysusingensemblecnntechniques AT sandhyaclement detectionofcovid19inchestxraysusingensemblecnntechniques |