COVID-19 detection in CT images with deep learning: A voting-based scheme and cross-datasets analysis
Early detection and diagnosis are critical factors to control the COVID-19 spreading. A number of deep learning-based methodologies have been recently proposed for COVID-19 screening in CT scans as a tool to automate and help with the diagnosis. These approaches, however, suffer from at least one of...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2020-01-01
|
| Series: | Informatics in Medicine Unlocked |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352914820305773 |
| _version_ | 1818203806328946688 |
|---|---|
| author | Pedro Silva Eduardo Luz Guilherme Silva Gladston Moreira Rodrigo Silva Diego Lucio David Menotti |
| author_facet | Pedro Silva Eduardo Luz Guilherme Silva Gladston Moreira Rodrigo Silva Diego Lucio David Menotti |
| author_sort | Pedro Silva |
| collection | DOAJ |
| description | Early detection and diagnosis are critical factors to control the COVID-19 spreading. A number of deep learning-based methodologies have been recently proposed for COVID-19 screening in CT scans as a tool to automate and help with the diagnosis. These approaches, however, suffer from at least one of the following problems: (i) they treat each CT scan slice independently and (ii) the methods are trained and tested with sets of images from the same dataset. Treating the slices independently means that the same patient may appear in the training and test sets at the same time which may produce misleading results. It also raises the question of whether the scans from the same patient should be evaluated as a group or not. Moreover, using a single dataset raises concerns about the generalization of the methods. Different datasets tend to present images of varying quality which may come from different types of CT machines reflecting the conditions of the countries and cities from where they come from. In order to address these two problems, in this work, we propose an Efficient Deep Learning Technique for the screening of COVID-19 with a voting-based approach. In this approach, the images from a given patient are classified as group in a voting system. The approach is tested in the two biggest datasets of COVID-19 CT analysis with a patient-based split. A cross dataset study is also presented to assess the robustness of the models in a more realistic scenario in which data comes from different distributions. The cross-dataset analysis has shown that the generalization power of deep learning models is far from acceptable for the task since accuracy drops from 87.68% to 56.16% on the best evaluation scenario. These results highlighted that the methods that aim at COVID-19 detection in CT-images have to improve significantly to be considered as a clinical option and larger and more diverse datasets are needed to evaluate the methods in a realistic scenario. |
| first_indexed | 2024-12-12T03:31:12Z |
| format | Article |
| id | doaj.art-5fa4ee3e8a604fdaa3a73b76e586a2fc |
| institution | Directory Open Access Journal |
| issn | 2352-9148 |
| language | English |
| last_indexed | 2024-12-12T03:31:12Z |
| publishDate | 2020-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Informatics in Medicine Unlocked |
| spelling | doaj.art-5fa4ee3e8a604fdaa3a73b76e586a2fc2022-12-22T00:39:55ZengElsevierInformatics in Medicine Unlocked2352-91482020-01-0120100427COVID-19 detection in CT images with deep learning: A voting-based scheme and cross-datasets analysisPedro Silva0Eduardo Luz1Guilherme Silva2Gladston Moreira3Rodrigo Silva4Diego Lucio5David Menotti6Computing Department, Universidade Federal de Ouro Preto (UFOP), MG, BrazilComputing Department, Universidade Federal de Ouro Preto (UFOP), MG, BrazilDepartment of Control and Automation Engineering, Universidade Federal de Ouro Preto (UFOP), MG, BrazilComputing Department, Universidade Federal de Ouro Preto (UFOP), MG, Brazil; Corresponding author.Computing Department, Universidade Federal de Ouro Preto (UFOP), MG, BrazilDepartment of Informatics, Universidade Federal do Parana (UFPR), PR, BrazilDepartment of Informatics, Universidade Federal do Parana (UFPR), PR, BrazilEarly detection and diagnosis are critical factors to control the COVID-19 spreading. A number of deep learning-based methodologies have been recently proposed for COVID-19 screening in CT scans as a tool to automate and help with the diagnosis. These approaches, however, suffer from at least one of the following problems: (i) they treat each CT scan slice independently and (ii) the methods are trained and tested with sets of images from the same dataset. Treating the slices independently means that the same patient may appear in the training and test sets at the same time which may produce misleading results. It also raises the question of whether the scans from the same patient should be evaluated as a group or not. Moreover, using a single dataset raises concerns about the generalization of the methods. Different datasets tend to present images of varying quality which may come from different types of CT machines reflecting the conditions of the countries and cities from where they come from. In order to address these two problems, in this work, we propose an Efficient Deep Learning Technique for the screening of COVID-19 with a voting-based approach. In this approach, the images from a given patient are classified as group in a voting system. The approach is tested in the two biggest datasets of COVID-19 CT analysis with a patient-based split. A cross dataset study is also presented to assess the robustness of the models in a more realistic scenario in which data comes from different distributions. The cross-dataset analysis has shown that the generalization power of deep learning models is far from acceptable for the task since accuracy drops from 87.68% to 56.16% on the best evaluation scenario. These results highlighted that the methods that aim at COVID-19 detection in CT-images have to improve significantly to be considered as a clinical option and larger and more diverse datasets are needed to evaluate the methods in a realistic scenario.http://www.sciencedirect.com/science/article/pii/S2352914820305773COVID-19Deep learningEfficientNetPneumoniaChest radiography |
| spellingShingle | Pedro Silva Eduardo Luz Guilherme Silva Gladston Moreira Rodrigo Silva Diego Lucio David Menotti COVID-19 detection in CT images with deep learning: A voting-based scheme and cross-datasets analysis Informatics in Medicine Unlocked COVID-19 Deep learning EfficientNet Pneumonia Chest radiography |
| title | COVID-19 detection in CT images with deep learning: A voting-based scheme and cross-datasets analysis |
| title_full | COVID-19 detection in CT images with deep learning: A voting-based scheme and cross-datasets analysis |
| title_fullStr | COVID-19 detection in CT images with deep learning: A voting-based scheme and cross-datasets analysis |
| title_full_unstemmed | COVID-19 detection in CT images with deep learning: A voting-based scheme and cross-datasets analysis |
| title_short | COVID-19 detection in CT images with deep learning: A voting-based scheme and cross-datasets analysis |
| title_sort | covid 19 detection in ct images with deep learning a voting based scheme and cross datasets analysis |
| topic | COVID-19 Deep learning EfficientNet Pneumonia Chest radiography |
| url | http://www.sciencedirect.com/science/article/pii/S2352914820305773 |
| work_keys_str_mv | AT pedrosilva covid19detectioninctimageswithdeeplearningavotingbasedschemeandcrossdatasetsanalysis AT eduardoluz covid19detectioninctimageswithdeeplearningavotingbasedschemeandcrossdatasetsanalysis AT guilhermesilva covid19detectioninctimageswithdeeplearningavotingbasedschemeandcrossdatasetsanalysis AT gladstonmoreira covid19detectioninctimageswithdeeplearningavotingbasedschemeandcrossdatasetsanalysis AT rodrigosilva covid19detectioninctimageswithdeeplearningavotingbasedschemeandcrossdatasetsanalysis AT diegolucio covid19detectioninctimageswithdeeplearningavotingbasedschemeandcrossdatasetsanalysis AT davidmenotti covid19detectioninctimageswithdeeplearningavotingbasedschemeandcrossdatasetsanalysis |