2021
DOI: 10.1101/2021.12.07.21267367
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Diagnosis of COVID-19 Using CT image Radiomics Features: A Comprehensive Machine Learning Study Involving 26,307 Patients

Abstract: Purpose: To derive and validate an effective radiomics-based model for differentiation of COVID-19 pneumonia from other lung diseases using a very large cohort of patients. Methods: We collected 19 private and 5 public datasets, accumulating to 26,307 individual patient images (15,148 COVID-19; 9,657 with other lung diseases e.g. non-COVID-19 pneumonia, lung cancer, pulmonary embolism; 1502 normal cases). Images were automatically segmented using a validated deep learning (DL) model and the results carefully … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
8
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
4
2

Relationship

4
2

Authors

Journals

citations
Cited by 10 publications
(8 citation statements)
references
References 67 publications
0
8
0
Order By: Relevance
“…For the feature selection technique RFE and the RF classifier, they obtained AUC=1.00. A Recent study by Shiri et al [22] indicated without the use of any other diagnostic test, CT-based radiomics derived from lung, paired with ML method, can enable highly effective identification of COVID-19.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…For the feature selection technique RFE and the RF classifier, they obtained AUC=1.00. A Recent study by Shiri et al [22] indicated without the use of any other diagnostic test, CT-based radiomics derived from lung, paired with ML method, can enable highly effective identification of COVID-19.…”
Section: Discussionmentioning
confidence: 99%
“…Subsequently, for selection of the most relevant features, 5 different feature selection algorithms including Correlation attribute evaluation, information gain attribute, wrapper subset feature selection, Relief method, and correlation-based feature selection using WEKA (version 3.8.2) were deployed. Unlike other studies [21, 22], the most pertinent features were finally selected using voting method by evaluating the performances of algorithms. We have chosen features that were frequent among more than 4 algorithms in this performance assessment stage.…”
Section: Methodsmentioning
confidence: 99%
“…The field of radiomics opens pathways for the study of normal tissues, cancer, and many other diseases, including potentially the newly emerging COVID-19 disease 6,7,29,36-40 . Specifically, Xie et al 41 evaluated the potential of a radiomics framework to diagnose COVID-19 from CT images.…”
Section: Introductionmentioning
confidence: 99%
“…CT aids in the diagnosis and management of COVID-19 patients and could be potentially used as an outcome/survival prediction tool, towards enhanced treatment planning [5][6][7] . CT scanning has been utilized as a highly sensitive tool for COVID-19 diagnosis 8 since it is fast and generates quantifiable features (e.g., the extent to which lung lobes are involved) and nonquantifiable features (e.g., ground-glass opacities and their laterality) to assess COVID-19 pneumonia, besides the enhanced sensitivity compared to RT-PCR 9 .…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation