Treffer: Automated machine learning for the identification of asymptomatic COVID-19 carriers based on chest CT images.

Title:
Automated machine learning for the identification of asymptomatic COVID-19 carriers based on chest CT images.
Authors:
Yin M; Department of Gastroenterology, The First Affiliated Hospital of Soochow University, 215006, Suzhou, Jiangsu, China.; Suzhou Clinical Center of Digestive Diseases, 215006, Suzhou, Jiangsu, China., Xu C; Department of Radiotherapy, The First Affiliated Hospital of Soochow University, 215006, Suzhou, Jiangsu, China., Zhu J; Department of Gastroenterology, The First Affiliated Hospital of Soochow University, 215006, Suzhou, Jiangsu, China.; The 23th ward, Yangzhou Third People's Hospital, 225000, Yangzhou, Jiangsu, China.; Suzhou Clinical Center of Digestive Diseases, 215006, Suzhou, Jiangsu, China., Xue Y; Medical School, Soochow University, 215006, Suzhou, Jiangsu, China., Zhou Y; Medical School, Soochow University, 215006, Suzhou, Jiangsu, China., He Y; Medical School, Soochow University, 215006, Suzhou, Jiangsu, China., Lin J; Department of Gastroenterology, The First Affiliated Hospital of Soochow University, 215006, Suzhou, Jiangsu, China.; Suzhou Clinical Center of Digestive Diseases, 215006, Suzhou, Jiangsu, China., Liu L; Department of Gastroenterology, The First Affiliated Hospital of Soochow University, 215006, Suzhou, Jiangsu, China.; Suzhou Clinical Center of Digestive Diseases, 215006, Suzhou, Jiangsu, China., Gao J; Department of Gastroenterology, The First Affiliated Hospital of Soochow University, 215006, Suzhou, Jiangsu, China.; Suzhou Clinical Center of Digestive Diseases, 215006, Suzhou, Jiangsu, China., Liu X; Department of Gastroenterology, The First Affiliated Hospital of Soochow University, 215006, Suzhou, Jiangsu, China.; Suzhou Clinical Center of Digestive Diseases, 215006, Suzhou, Jiangsu, China., Shen D; Department of Respiratory Medicine, The First Affiliated Hospital of Soochow University, 215006, Suzhou, Jiangsu, China. shendan@suda.edu.cn., Fu C; Department of Respiratory Medicine, The First Affiliated Hospital of Soochow University, 215006, Suzhou, Jiangsu, China. fucuipingjy@163.com.
Source:
BMC medical imaging [BMC Med Imaging] 2024 Feb 27; Vol. 24 (1), pp. 50. Date of Electronic Publication: 2024 Feb 27.
Publication Type:
Journal Article; Research Support, Non-U.S. Gov't
Language:
English
Journal Info:
Publisher: BioMed Central Country of Publication: England NLM ID: 100968553 Publication Model: Electronic Cited Medium: Internet ISSN: 1471-2342 (Electronic) Linking ISSN: 14712342 NLM ISO Abbreviation: BMC Med Imaging Subsets: MEDLINE
Imprint Name(s):
Original Publication: London : BioMed Central, [2001-
MeSH Terms:
Deep Learning* , COVID-19*/diagnostic imaging, Humans ; Retrospective Studies ; Tomography, X-Ray Computed/methods ; Machine Learning
References:
N Engl J Med. 2020 Mar 26;382(13):1199-1207. (PMID: 31995857)
Chest. 2012 Jan;141(1):222-231. (PMID: 22215830)
JAMA. 2020 Mar 17;323(11):1061-1069. (PMID: 32031570)
AJR Am J Roentgenol. 2018 Feb;210(2):307-313. (PMID: 29140119)
QJM. 2021 Nov 5;114(7):496-501. (PMID: 34156436)
J Med Imaging (Bellingham). 2020 Jan;7(1):014504. (PMID: 32118090)
Comput Biol Med. 2020 Jun;121:103792. (PMID: 32568675)
Nat Commun. 2020 Aug 14;11(1):4080. (PMID: 32796848)
Sci Rep. 2021 Oct 12;11(1):20189. (PMID: 34642404)
Int J Mol Sci. 2022 Oct 05;23(19):. (PMID: 36233123)
AJR Am J Roentgenol. 2020 Jun;214(6):1287-1294. (PMID: 32134681)
J Microbiol Immunol Infect. 2021 Feb;54(1):12-16. (PMID: 32425996)
Diagnostics (Basel). 2023 May 10;13(10):. (PMID: 37238175)
J Med Internet Res. 2020 Nov 6;22(11):e24018. (PMID: 33027032)
Prog Biophys Mol Biol. 2023 May;179:1-9. (PMID: 36809830)
EBioMedicine. 2019 Dec;50:156-165. (PMID: 31735556)
Drug Discov Today. 2019 Mar;24(3):773-780. (PMID: 30472429)
Health Technol Assess. 2022 Nov 30;:. (PMID: 36562494)
Arch Pathol Lab Med. 2017 Jul;141(7):960-969. (PMID: 28467213)
JAMA. 2020 Feb 25;323(8):707-708. (PMID: 31971553)
Multimed Tools Appl. 2021;80(4):5423-5447. (PMID: 33041635)
Chaos Solitons Fractals. 2020 Nov;140:110153. (PMID: 32834641)
Antibiotics (Basel). 2020 Jan 31;9(2):. (PMID: 32023854)
Sci Rep. 2023 Feb 4;13(1):2040. (PMID: 36739358)
Eur J Radiol. 2020 May;126:108961. (PMID: 32229322)
J Clin Med. 2020 Jun 01;9(6):. (PMID: 32492874)
Artif Intell Med. 2020 Jan;102:101752. (PMID: 31980091)
J Digit Imaging. 2023 Jun;36(3):827-836. (PMID: 36596937)
Biomed Signal Process Control. 2023 May;83:104642. (PMID: 36818992)
Comput Biol Med. 2022 Oct;149:106070. (PMID: 36099862)
Diagnostics (Basel). 2023 May 11;13(10):. (PMID: 37238180)
Grant Information:
KJXW2019001 Youth Program of Suzhou Health Committee; 82100109 National Natural Science Foundation of China
Contributed Indexing:
Keywords: Asymptomatic; Automated machine learning; COVID-19; Prediction model
Entry Date(s):
Date Created: 20240227 Date Completed: 20240229 Latest Revision: 20250806
Update Code:
20250807
PubMed Central ID:
PMC10900643
DOI:
10.1186/s12880-024-01211-w
PMID:
38413923
Database:
MEDLINE

Weitere Informationen

Background: Asymptomatic COVID-19 carriers with normal chest computed tomography (CT) scans have perpetuated the ongoing pandemic of this disease. This retrospective study aimed to use automated machine learning (AutoML) to develop a prediction model based on CT characteristics for the identification of asymptomatic carriers.
Methods: Asymptomatic carriers were from Yangzhou Third People's Hospital from August 1st, 2020, to March 31st, 2021, and the control group included a healthy population from a nonepizootic area with two negative RT‒PCR results within 48 h. All CT images were preprocessed using MATLAB. Model development and validation were conducted in R with the H2O package. The models were built based on six algorithms, e.g., random forest and deep neural network (DNN), and a training set (n = 691). The models were improved by automatically adjusting hyperparameters for an internal validation set (n = 306). The performance of the obtained models was evaluated based on a dataset from Suzhou (n = 178) using the area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and F1 score.
Results: A total of 1,175 images were preprocessed with high stability. Six models were developed, and the performance of the DNN model ranked first, with an AUC value of 0.898 for the test set. The sensitivity, specificity, PPV, NPV, F1 score and accuracy of the DNN model were 0.820, 0.854, 0.849, 0.826, 0.834 and 0.837, respectively. A plot of a local interpretable model-agnostic explanation demonstrated how different variables worked in identifying asymptomatic carriers.
Conclusions: Our study demonstrates that AutoML models based on CT images can be used to identify asymptomatic carriers. The most promising model for clinical implementation is the DNN-algorithm-based model.
(© 2024. The Author(s).)

The authors declare no competing interests.