Inter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography
<i>Background</i>: For COVID-19 lung severity, segmentation of lungs on computed tomography (CT) is the first crucial step. Current deep learning (DL)-based Artificial Intelligence (AI) models have a bias in the training stage of segmentation because only one set of ground truth (GT) ann...
Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
MDPI AG
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/1dc2ceaec98142dea5fc58b63dc51b62 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:1dc2ceaec98142dea5fc58b63dc51b62 |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:1dc2ceaec98142dea5fc58b63dc51b622021-11-25T17:20:53ZInter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography10.3390/diagnostics111120252075-4418https://doaj.org/article/1dc2ceaec98142dea5fc58b63dc51b622021-11-01T00:00:00Zhttps://www.mdpi.com/2075-4418/11/11/2025https://doaj.org/toc/2075-4418<i>Background</i>: For COVID-19 lung severity, segmentation of lungs on computed tomography (CT) is the first crucial step. Current deep learning (DL)-based Artificial Intelligence (AI) models have a bias in the training stage of segmentation because only one set of ground truth (GT) annotations are evaluated. We propose a robust and stable inter-variability analysis of CT lung segmentation in COVID-19 to avoid the effect of bias. <i>Methodology</i>: The proposed inter-variability study consists of two GT tracers for lung segmentation on chest CT. Three AI models, PSP Net, VGG-SegNet, and ResNet-SegNet, were trained using GT annotations. We hypothesized that if AI models are trained on the GT tracings from multiple experience levels, and if the AI performance on the test data between these AI models is within the 5% range, one can consider such an AI model robust and unbiased. The K5 protocol (training to testing: 80%:20%) was adapted. Ten kinds of metrics were used for performance evaluation. <i>Results</i>: The database consisted of 5000 CT chest images from 72 COVID-19-infected patients. By computing the coefficient of correlations (CC) between the output of the two AI models trained corresponding to the two GT tracers, computing their differences in their CC, and repeating the process for all three AI-models, we show the differences as 0%, 0.51%, and 2.04% (all < 5%), thereby validating the hypothesis. The performance was comparable; however, it had the following order: ResNet-SegNet > PSP Net > VGG-SegNet. <i>Conclusions</i>: The AI models were clinically robust and stable during the inter-variability analysis on the CT lung segmentation on COVID-19 patients.Jasjit S. SuriSushant AgarwalPranav ElavarthiRajesh PathakVedmanvitha KetireddyMarta ColumbuLuca SabaSuneet K. GuptaGavino FaaInder M. SinghMonika TurkParamjit S. ChadhaAmer M. JohriNarendra N. KhannaKlaudija ViskovicSophie MavrogeniJohn R. LairdGyan PareekMartin MinerDavid W. SobelAntonella BalestrieriPetros P. SfikakisGeorge TsoulfasAthanasios ProtogerouDurga Prasanna MisraVikas AgarwalGeorge D. KitasJagjit S. TejiMustafa Al-MainiSurinder K. DhanjilAndrew NicolaidesAditya SharmaVijay RathoreMostafa FatemiAzra AlizadPudukode R. KrishnanNagy FerencZoltan RuzsaArchna GuptaSubbaram NaiduMannudeep K. KalraMDPI AGarticleCOVID-19computed tomographylungsvariabilitysegmentationhybrid deep learningMedicine (General)R5-920ENDiagnostics, Vol 11, Iss 2025, p 2025 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
COVID-19 computed tomography lungs variability segmentation hybrid deep learning Medicine (General) R5-920 |
| spellingShingle |
COVID-19 computed tomography lungs variability segmentation hybrid deep learning Medicine (General) R5-920 Jasjit S. Suri Sushant Agarwal Pranav Elavarthi Rajesh Pathak Vedmanvitha Ketireddy Marta Columbu Luca Saba Suneet K. Gupta Gavino Faa Inder M. Singh Monika Turk Paramjit S. Chadha Amer M. Johri Narendra N. Khanna Klaudija Viskovic Sophie Mavrogeni John R. Laird Gyan Pareek Martin Miner David W. Sobel Antonella Balestrieri Petros P. Sfikakis George Tsoulfas Athanasios Protogerou Durga Prasanna Misra Vikas Agarwal George D. Kitas Jagjit S. Teji Mustafa Al-Maini Surinder K. Dhanjil Andrew Nicolaides Aditya Sharma Vijay Rathore Mostafa Fatemi Azra Alizad Pudukode R. Krishnan Nagy Ferenc Zoltan Ruzsa Archna Gupta Subbaram Naidu Mannudeep K. Kalra Inter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography |
| description |
<i>Background</i>: For COVID-19 lung severity, segmentation of lungs on computed tomography (CT) is the first crucial step. Current deep learning (DL)-based Artificial Intelligence (AI) models have a bias in the training stage of segmentation because only one set of ground truth (GT) annotations are evaluated. We propose a robust and stable inter-variability analysis of CT lung segmentation in COVID-19 to avoid the effect of bias. <i>Methodology</i>: The proposed inter-variability study consists of two GT tracers for lung segmentation on chest CT. Three AI models, PSP Net, VGG-SegNet, and ResNet-SegNet, were trained using GT annotations. We hypothesized that if AI models are trained on the GT tracings from multiple experience levels, and if the AI performance on the test data between these AI models is within the 5% range, one can consider such an AI model robust and unbiased. The K5 protocol (training to testing: 80%:20%) was adapted. Ten kinds of metrics were used for performance evaluation. <i>Results</i>: The database consisted of 5000 CT chest images from 72 COVID-19-infected patients. By computing the coefficient of correlations (CC) between the output of the two AI models trained corresponding to the two GT tracers, computing their differences in their CC, and repeating the process for all three AI-models, we show the differences as 0%, 0.51%, and 2.04% (all < 5%), thereby validating the hypothesis. The performance was comparable; however, it had the following order: ResNet-SegNet > PSP Net > VGG-SegNet. <i>Conclusions</i>: The AI models were clinically robust and stable during the inter-variability analysis on the CT lung segmentation on COVID-19 patients. |
| format |
article |
| author |
Jasjit S. Suri Sushant Agarwal Pranav Elavarthi Rajesh Pathak Vedmanvitha Ketireddy Marta Columbu Luca Saba Suneet K. Gupta Gavino Faa Inder M. Singh Monika Turk Paramjit S. Chadha Amer M. Johri Narendra N. Khanna Klaudija Viskovic Sophie Mavrogeni John R. Laird Gyan Pareek Martin Miner David W. Sobel Antonella Balestrieri Petros P. Sfikakis George Tsoulfas Athanasios Protogerou Durga Prasanna Misra Vikas Agarwal George D. Kitas Jagjit S. Teji Mustafa Al-Maini Surinder K. Dhanjil Andrew Nicolaides Aditya Sharma Vijay Rathore Mostafa Fatemi Azra Alizad Pudukode R. Krishnan Nagy Ferenc Zoltan Ruzsa Archna Gupta Subbaram Naidu Mannudeep K. Kalra |
| author_facet |
Jasjit S. Suri Sushant Agarwal Pranav Elavarthi Rajesh Pathak Vedmanvitha Ketireddy Marta Columbu Luca Saba Suneet K. Gupta Gavino Faa Inder M. Singh Monika Turk Paramjit S. Chadha Amer M. Johri Narendra N. Khanna Klaudija Viskovic Sophie Mavrogeni John R. Laird Gyan Pareek Martin Miner David W. Sobel Antonella Balestrieri Petros P. Sfikakis George Tsoulfas Athanasios Protogerou Durga Prasanna Misra Vikas Agarwal George D. Kitas Jagjit S. Teji Mustafa Al-Maini Surinder K. Dhanjil Andrew Nicolaides Aditya Sharma Vijay Rathore Mostafa Fatemi Azra Alizad Pudukode R. Krishnan Nagy Ferenc Zoltan Ruzsa Archna Gupta Subbaram Naidu Mannudeep K. Kalra |
| author_sort |
Jasjit S. Suri |
| title |
Inter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography |
| title_short |
Inter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography |
| title_full |
Inter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography |
| title_fullStr |
Inter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography |
| title_full_unstemmed |
Inter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography |
| title_sort |
inter-variability study of covlias 1.0: hybrid deep learning models for covid-19 lung segmentation in computed tomography |
| publisher |
MDPI AG |
| publishDate |
2021 |
| url |
https://doaj.org/article/1dc2ceaec98142dea5fc58b63dc51b62 |
| work_keys_str_mv |
AT jasjitssuri intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT sushantagarwal intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT pranavelavarthi intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT rajeshpathak intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT vedmanvithaketireddy intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT martacolumbu intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT lucasaba intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT suneetkgupta intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT gavinofaa intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT indermsingh intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT monikaturk intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT paramjitschadha intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT amermjohri intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT narendrankhanna intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT klaudijaviskovic intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT sophiemavrogeni intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT johnrlaird intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT gyanpareek intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT martinminer intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT davidwsobel intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT antonellabalestrieri intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT petrospsfikakis intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT georgetsoulfas intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT athanasiosprotogerou intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT durgaprasannamisra intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT vikasagarwal intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT georgedkitas intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT jagjitsteji intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT mustafaalmaini intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT surinderkdhanjil intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT andrewnicolaides intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT adityasharma intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT vijayrathore intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT mostafafatemi intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT azraalizad intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT pudukoderkrishnan intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT nagyferenc intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT zoltanruzsa intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT archnagupta intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT subbaramnaidu intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography AT mannudeepkkalra intervariabilitystudyofcovlias10hybriddeeplearningmodelsforcovid19lungsegmentationincomputedtomography |
| _version_ |
1718412455464402944 |