Am J Transl Res. 2021 Jun 15;13(6):6166-6174. eCollection 2021.
Development of a random forest model to classify sarcoidosis and tuberculosis.
American journal of translational research
Jun Ma, Hongyun Yin, Xiaohui Hao, Wei Sha, Haiyan Cui
Affiliations
Affiliations
- Department of Tuberculosis and Shanghai Key Lab of Tuberculosis, Shanghai Pulmonary Hospital Affiliated to Tongji University School of Medicine 507 Zhengmin Road, Shanghai 200433, China.
- Clinic and Research Center of Tuberculosis, Shanghai Key Lab of Tuberculosis, Shanghai Pulmonary Hospital, Tongji University School of Medicine 507 Zhengmin Road, Shanghai 200433, China.
PMID: 34306355
PMCID: PMC8290672
Abstract
OBJECTIVE: To identify significant diagnostic factors and establish a predictive model for diagnosis of sarcoidosis and tuberculosis.
METHODS: This study included 252 patients (123 cases of lung sarcoidosis and 129 cases of lung tuberculosis) who underwent laboratory evaluation, including routine hematologic testing, serum immunology, blood coagulation, angiotensin-converting enzyme, and T lymphocyte subset. The factors that statistically different between the two groups were identified by an independent sample t test first, and then processed by the random forest model to distinguish two diseases with the classification function. Moreover, the diagnostic performance of the predictive random forest model was evaluated through the identification of individual contribution of various diagnostic factors conducted by the model.
RESULTS: The random forest model revealed a classification error rate of 24.9%. Among all of the statistically significant diagnostic factors, the individual factors with the greatest and second contribution were angiotensin-converting enzyme and prothrombin time, respectively. The area under the receiver operating characteristic (ROC) curve of the random forest prediction model was 0.915.
CONCLUSION: The random forest model can be used to distinguish between sarcoidosis and tuberculosis by incorporating statistically significant diagnostic factors, which is of potential clinical application value.
AJTR Copyright © 2021.
Keywords: ACE; Random forest model; differential diagnosis; sarcoidosis; tuberculosis
Conflict of interest statement
None.
References
- Clin Chest Med. 2015 Dec;36(4):631-41 - PubMed
- Clin Rev Allergy Immunol. 2015 Aug;49(1):54-62 - PubMed
- Trends Microbiol. 2017 Aug;25(8):688-697 - PubMed
- Clin Chem Lab Med. 2018 Jun 27;56(7):1117-1125 - PubMed
- Sci Rep. 2016 Oct 13;6:34913 - PubMed
- Bioinformation. 2020 Jul 31;16(7):539-546 - PubMed
- Clin Infect Dis. 2017 Jan 15;64(2):e1-e33 - PubMed
- Arkh Patol. 2001 Jan-Feb;63(1):6-11 - PubMed
- Ann Thorac Med. 2014 Oct;9(4):232-5 - PubMed
- Expert Rev Clin Pharmacol. 2018 Jul;11(7):677-687 - PubMed
- Clin Biochem. 2018 Sep;59:1-8 - PubMed
- Respir Med. 2011 Sep;105(9):1263-7 - PubMed
- J Infect. 2015 Apr;70(4):324-34 - PubMed
- Sarcoidosis Vasc Diffuse Lung Dis. 2020;36(3):209-216 - PubMed
- Postgrad Med. 2017 Jan;129(1):149-158 - PubMed
- Infect Dis Poverty. 2018 Oct 20;7(1):106 - PubMed
- Eur Respir Rev. 2018 Feb 28;27(147): - PubMed
- Ocul Immunol Inflamm. 2019;27(7):1041-1048 - PubMed
- Clin Lab. 2018 Jan 1;64(1):135-140 - PubMed
- Tuberculosis (Edinb). 2015 Sep;95(5):527-31 - PubMed
- J Cytol. 2019 Apr-Jun;36(2):128-130 - PubMed
- Indian J Tuberc. 2017 Oct;64(4):243-245 - PubMed
- Medicine (Baltimore). 2016 May;95(19):e3579 - PubMed
- Eur Respir Rev. 2017 Aug 9;26(145): - PubMed
- Int Immunopharmacol. 2015 Mar;25(1):174-9 - PubMed
- Front Immunol. 2018 Sep 19;9:2108 - PubMed
- Ann Am Thorac Soc. 2017 Dec;14(Supplement_6):S415-S420 - PubMed
- Sarcoidosis Vasc Diffuse Lung Dis. 1999 Sep;16(2):149-73 - PubMed
- BMC Ophthalmol. 2016 Feb 16;16:19 - PubMed
- Br J Haematol. 1996 Jun;93(4):943-9 - PubMed
- PLoS Med. 2005 Dec;2(12):e381 - PubMed
- J Fr Ophtalmol. 2018 Dec;41(10):e451-e467 - PubMed
Publication Types