A Few-Shot U-Net Deep Learning Model for COVID-19 Infected Area Segmentation in CT Images
Recent studies indicate that detecting radiographic patterns on CT chest scans can yield high sensitivity and specificity for COVID-19 identification. In this paper, we scrutinize the effectiveness of deep learning models for semantic segmentation of pneumonia-infected area segmentation in CT images...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/6/2215 |
| _version_ | 1827697067571544064 |
|---|---|
| author | Athanasios Voulodimos Eftychios Protopapadakis Iason Katsamenis Anastasios Doulamis Nikolaos Doulamis |
| author_facet | Athanasios Voulodimos Eftychios Protopapadakis Iason Katsamenis Anastasios Doulamis Nikolaos Doulamis |
| author_sort | Athanasios Voulodimos |
| collection | DOAJ |
| description | Recent studies indicate that detecting radiographic patterns on CT chest scans can yield high sensitivity and specificity for COVID-19 identification. In this paper, we scrutinize the effectiveness of deep learning models for semantic segmentation of pneumonia-infected area segmentation in CT images for the detection of COVID-19. Traditional methods for CT scan segmentation exploit a supervised learning paradigm, so they (a) require large volumes of data for their training, and (b) assume fixed (static) network weights once the training procedure has been completed. Recently, to overcome these difficulties, few-shot learning (FSL) has been introduced as a general concept of network model training using a very small amount of samples. In this paper, we explore the efficacy of few-shot learning in U-Net architectures, allowing for a dynamic fine-tuning of the network weights as new few samples are being fed into the U-Net. Experimental results indicate improvement in the segmentation accuracy of identifying COVID-19 infected regions. In particular, using 4-fold cross-validation results of the different classifiers, we observed an improvement of 5.388 ± 3.046% for all test data regarding the IoU metric and a similar increment of 5.394 ± 3.015% for the F1 score. Moreover, the statistical significance of the improvement obtained using our proposed few-shot U-Net architecture compared with the traditional U-Net model was confirmed by applying the Kruskal-Wallis test (<i>p</i>-value = 0.026). |
| first_indexed | 2024-03-10T13:01:04Z |
| format | Article |
| id | doaj.art-7fdcf674bba5437198f57299bdfe5dd0 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T13:01:04Z |
| publishDate | 2021-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-7fdcf674bba5437198f57299bdfe5dd02023-11-21T11:32:19ZengMDPI AGSensors1424-82202021-03-01216221510.3390/s21062215A Few-Shot U-Net Deep Learning Model for COVID-19 Infected Area Segmentation in CT ImagesAthanasios Voulodimos0Eftychios Protopapadakis1Iason Katsamenis2Anastasios Doulamis3Nikolaos Doulamis4Department of Informatics and Computer Engineering, University of West Attica, 12243 Athens, GreeceDepartment of Informatics and Computer Engineering, University of West Attica, 12243 Athens, GreeceSchool of Rural and Surveying Engineering, National Technical University of Athens, 15780 Athens, GreeceSchool of Rural and Surveying Engineering, National Technical University of Athens, 15780 Athens, GreeceSchool of Rural and Surveying Engineering, National Technical University of Athens, 15780 Athens, GreeceRecent studies indicate that detecting radiographic patterns on CT chest scans can yield high sensitivity and specificity for COVID-19 identification. In this paper, we scrutinize the effectiveness of deep learning models for semantic segmentation of pneumonia-infected area segmentation in CT images for the detection of COVID-19. Traditional methods for CT scan segmentation exploit a supervised learning paradigm, so they (a) require large volumes of data for their training, and (b) assume fixed (static) network weights once the training procedure has been completed. Recently, to overcome these difficulties, few-shot learning (FSL) has been introduced as a general concept of network model training using a very small amount of samples. In this paper, we explore the efficacy of few-shot learning in U-Net architectures, allowing for a dynamic fine-tuning of the network weights as new few samples are being fed into the U-Net. Experimental results indicate improvement in the segmentation accuracy of identifying COVID-19 infected regions. In particular, using 4-fold cross-validation results of the different classifiers, we observed an improvement of 5.388 ± 3.046% for all test data regarding the IoU metric and a similar increment of 5.394 ± 3.015% for the F1 score. Moreover, the statistical significance of the improvement obtained using our proposed few-shot U-Net architecture compared with the traditional U-Net model was confirmed by applying the Kruskal-Wallis test (<i>p</i>-value = 0.026).https://www.mdpi.com/1424-8220/21/6/2215deep learningfew-shot learningsemantic segmentationCT imagesCOVID-19 |
| spellingShingle | Athanasios Voulodimos Eftychios Protopapadakis Iason Katsamenis Anastasios Doulamis Nikolaos Doulamis A Few-Shot U-Net Deep Learning Model for COVID-19 Infected Area Segmentation in CT Images Sensors deep learning few-shot learning semantic segmentation CT images COVID-19 |
| title | A Few-Shot U-Net Deep Learning Model for COVID-19 Infected Area Segmentation in CT Images |
| title_full | A Few-Shot U-Net Deep Learning Model for COVID-19 Infected Area Segmentation in CT Images |
| title_fullStr | A Few-Shot U-Net Deep Learning Model for COVID-19 Infected Area Segmentation in CT Images |
| title_full_unstemmed | A Few-Shot U-Net Deep Learning Model for COVID-19 Infected Area Segmentation in CT Images |
| title_short | A Few-Shot U-Net Deep Learning Model for COVID-19 Infected Area Segmentation in CT Images |
| title_sort | few shot u net deep learning model for covid 19 infected area segmentation in ct images |
| topic | deep learning few-shot learning semantic segmentation CT images COVID-19 |
| url | https://www.mdpi.com/1424-8220/21/6/2215 |
| work_keys_str_mv | AT athanasiosvoulodimos afewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages AT eftychiosprotopapadakis afewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages AT iasonkatsamenis afewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages AT anastasiosdoulamis afewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages AT nikolaosdoulamis afewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages AT athanasiosvoulodimos fewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages AT eftychiosprotopapadakis fewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages AT iasonkatsamenis fewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages AT anastasiosdoulamis fewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages AT nikolaosdoulamis fewshotunetdeeplearningmodelforcovid19infectedareasegmentationinctimages |