Reliability of Machine Learning in Eliminating Data Redundancy of Radiomics and Reflecting Pathophysiology in COVID-19 Pneumonia: Impact of CT Reconstruction Kernels on Accuracy

Yauhen Statsenko; Tetiana Habuza; Tatsiana Talako; Tetiana Kurbatova; Gillian Lylian Simiyu; Darya Smetanina; Juana Sido; Dana Sharif Qandil; Sarah Meribout; Juri G. Gelovani; Klaus Neidl Van Gorkom; Taleb M. Almansoori; Fatmah Al Zahmi; Tom Loney; Anthony Bedson; Nerissa Naidoo; Alireza Dehdashtian; Milos Ljubisavljevic; Jamal Al Koteesh; Karuna M. Das

doi:10.1109/ACCESS.2022.3211080

Reliability of Machine Learning in Eliminating Data Redundancy of Radiomics and Reflecting Pathophysiology in COVID-19 Pneumonia: Impact of CT Reconstruction Kernels on Accuracy

Yauhen Statsenko, Tetiana Habuza, Tatsiana Talako, Tetiana Kurbatova, Gillian Lylian Simiyu, Darya Smetanina, Juana Sido, Dana Sharif Qandil, Sarah Meribout, Juri G. Gelovani, Klaus Neidl Van Gorkom, Taleb M. Almansoori, Fatmah Al Zahmi, Tom Loney, Anthony Bedson, Nerissa Naidoo, Alireza Dehdashtian, Milos Ljubisavljevic, Jamal Al Koteesh, Karuna M. Das

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

Abstract

Background: Radiomical data are redundant but they might serve as a tool for lung quantitative assessment reflecting disease severity and actual physiological status of COVID-19 patients. Objective: Test the effectiveness of machine learning in eliminating data redundancy of radiomics and reflecting pathophysiologic changes in patients with COVID-19 pneumonia. Methods: We analyzed 605 cases admitted to Al Ain Hospital from 24 February to 1 July, 2020. They met the following inclusion criteria: age ≥ 18 years; inpatient admission; PCR positive for SARS-CoV-2; lung CT available at PACS. We categorized cases into 4 classes: mild <5% of pulmonary parenchymal involvement, moderate - 5-24%, severe - 25-49%, and critical ≥50 %. We used CT scans to build regression models predicting the oxygenation level, respiratory and cardiovascular functioning. Results: Radiomical findings are a reliable source of information to assess the functional status of patients with COVID-19.

Original language	English
Pages (from-to)	120901-120921
Number of pages	21
Journal	IEEE Access
Volume	10
DOIs	https://doi.org/10.1109/ACCESS.2022.3211080
Publication status	Published - 2022

Keywords

Blended machine learning model
COVID-19
SARC-CoV-2
functional outcomes
lung structural changes
pneumonia
radiomics
structure-function association

ASJC Scopus subject areas

Computer Science(all)
Materials Science(all)
Engineering(all)
Electrical and Electronic Engineering

Access to Document

10.1109/ACCESS.2022.3211080

Cite this

Statsenko, Y., Habuza, T., Talako, T., Kurbatova, T., Simiyu, G. L., Smetanina, D., Sido, J., Qandil, D. S., Meribout, S., Gelovani, J. G., Gorkom, K. N. V., Almansoori, T. M., Zahmi, F. A., Loney, T., Bedson, A., Naidoo, N., Dehdashtian, A., Ljubisavljevic, M., Koteesh, J. A., & Das, K. M. (2022). Reliability of Machine Learning in Eliminating Data Redundancy of Radiomics and Reflecting Pathophysiology in COVID-19 Pneumonia: Impact of CT Reconstruction Kernels on Accuracy. IEEE Access, 10, 120901-120921. https://doi.org/10.1109/ACCESS.2022.3211080

Statsenko, Y, Habuza, T, Talako, T, Kurbatova, T, Simiyu, GL, Smetanina, D, Sido, J, Qandil, DS, Meribout, S, Gelovani, JG, Gorkom, KNV, Almansoori, TM, Zahmi, FA, Loney, T, Bedson, A, Naidoo, N, Dehdashtian, A, Ljubisavljevic, M, Koteesh, JA & Das, KM 2022, 'Reliability of Machine Learning in Eliminating Data Redundancy of Radiomics and Reflecting Pathophysiology in COVID-19 Pneumonia: Impact of CT Reconstruction Kernels on Accuracy', IEEE Access, vol. 10, pp. 120901-120921. https://doi.org/10.1109/ACCESS.2022.3211080

@article{23605c7bdcef4e598a35d09daa248ce0,

title = "Reliability of Machine Learning in Eliminating Data Redundancy of Radiomics and Reflecting Pathophysiology in COVID-19 Pneumonia: Impact of CT Reconstruction Kernels on Accuracy",

abstract = "Background: Radiomical data are redundant but they might serve as a tool for lung quantitative assessment reflecting disease severity and actual physiological status of COVID-19 patients. Objective: Test the effectiveness of machine learning in eliminating data redundancy of radiomics and reflecting pathophysiologic changes in patients with COVID-19 pneumonia. Methods: We analyzed 605 cases admitted to Al Ain Hospital from 24 February to 1 July, 2020. They met the following inclusion criteria: age ≥ 18 years; inpatient admission; PCR positive for SARS-CoV-2; lung CT available at PACS. We categorized cases into 4 classes: mild <5% of pulmonary parenchymal involvement, moderate - 5-24%, severe - 25-49%, and critical ≥50 %. We used CT scans to build regression models predicting the oxygenation level, respiratory and cardiovascular functioning. Results: Radiomical findings are a reliable source of information to assess the functional status of patients with COVID-19.",

keywords = "Blended machine learning model, COVID-19, SARC-CoV-2, functional outcomes, lung structural changes, pneumonia, radiomics, structure-function association",

author = "Yauhen Statsenko and Tetiana Habuza and Tatsiana Talako and Tetiana Kurbatova and Simiyu, {Gillian Lylian} and Darya Smetanina and Juana Sido and Qandil, {Dana Sharif} and Sarah Meribout and Gelovani, {Juri G.} and Gorkom, {Klaus Neidl Van} and Almansoori, {Taleb M.} and Zahmi, {Fatmah Al} and Tom Loney and Anthony Bedson and Nerissa Naidoo and Alireza Dehdashtian and Milos Ljubisavljevic and Koteesh, {Jamal Al} and Das, {Karuna M.}",

note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",

year = "2022",

doi = "10.1109/ACCESS.2022.3211080",

language = "English",

volume = "10",

pages = "120901--120921",

journal = "IEEE Access",

issn = "2169-3536",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Reliability of Machine Learning in Eliminating Data Redundancy of Radiomics and Reflecting Pathophysiology in COVID-19 Pneumonia

T2 - Impact of CT Reconstruction Kernels on Accuracy

AU - Statsenko, Yauhen

AU - Habuza, Tetiana

AU - Talako, Tatsiana

AU - Kurbatova, Tetiana

AU - Simiyu, Gillian Lylian

AU - Smetanina, Darya

AU - Sido, Juana

AU - Qandil, Dana Sharif

AU - Meribout, Sarah

AU - Gelovani, Juri G.

AU - Gorkom, Klaus Neidl Van

AU - Almansoori, Taleb M.

AU - Zahmi, Fatmah Al

AU - Loney, Tom

AU - Bedson, Anthony

AU - Naidoo, Nerissa

AU - Dehdashtian, Alireza

AU - Ljubisavljevic, Milos

AU - Koteesh, Jamal Al

AU - Das, Karuna M.

PY - 2022

Y1 - 2022

N2 - Background: Radiomical data are redundant but they might serve as a tool for lung quantitative assessment reflecting disease severity and actual physiological status of COVID-19 patients. Objective: Test the effectiveness of machine learning in eliminating data redundancy of radiomics and reflecting pathophysiologic changes in patients with COVID-19 pneumonia. Methods: We analyzed 605 cases admitted to Al Ain Hospital from 24 February to 1 July, 2020. They met the following inclusion criteria: age ≥ 18 years; inpatient admission; PCR positive for SARS-CoV-2; lung CT available at PACS. We categorized cases into 4 classes: mild <5% of pulmonary parenchymal involvement, moderate - 5-24%, severe - 25-49%, and critical ≥50 %. We used CT scans to build regression models predicting the oxygenation level, respiratory and cardiovascular functioning. Results: Radiomical findings are a reliable source of information to assess the functional status of patients with COVID-19.

AB - Background: Radiomical data are redundant but they might serve as a tool for lung quantitative assessment reflecting disease severity and actual physiological status of COVID-19 patients. Objective: Test the effectiveness of machine learning in eliminating data redundancy of radiomics and reflecting pathophysiologic changes in patients with COVID-19 pneumonia. Methods: We analyzed 605 cases admitted to Al Ain Hospital from 24 February to 1 July, 2020. They met the following inclusion criteria: age ≥ 18 years; inpatient admission; PCR positive for SARS-CoV-2; lung CT available at PACS. We categorized cases into 4 classes: mild <5% of pulmonary parenchymal involvement, moderate - 5-24%, severe - 25-49%, and critical ≥50 %. We used CT scans to build regression models predicting the oxygenation level, respiratory and cardiovascular functioning. Results: Radiomical findings are a reliable source of information to assess the functional status of patients with COVID-19.

KW - Blended machine learning model

KW - COVID-19

KW - SARC-CoV-2

KW - functional outcomes

KW - lung structural changes

KW - pneumonia

KW - radiomics

KW - structure-function association

UR - http://www.scopus.com/inward/record.url?scp=85139444789&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85139444789&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2022.3211080

DO - 10.1109/ACCESS.2022.3211080

M3 - Article

AN - SCOPUS:85139444789

SN - 2169-3536

VL - 10

SP - 120901

EP - 120921

JO - IEEE Access

JF - IEEE Access

ER -

Reliability of Machine Learning in Eliminating Data Redundancy of Radiomics and Reflecting Pathophysiology in COVID-19 Pneumonia: Impact of CT Reconstruction Kernels on Accuracy

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this