Assessing Diagnostic Accuracy of Doctors Without a Gold Standard using Bayesian Networks and Kmodes Dlustering Algorithm
Background & Aim: The diagnostic accuracy of a test is the ability to discriminate accuratelybetween patients who have and do not have the target disease. A common problem in assessing thediagnostic accuracy of doctors is the unknown true disease status which in the literature is referredas “ab...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Tehran University of Medical Sciences
2019-03-01
|
Series: | Journal of Biostatistics and Epidemiology |
Subjects: | |
Online Access: | https://jbe.tums.ac.ir/index.php/jbe/article/view/226 |
id |
doaj-d18d06636a4f45ce8c2da19f7101364f |
---|---|
record_format |
Article |
spelling |
doaj-d18d06636a4f45ce8c2da19f7101364f2020-12-06T04:14:38ZengTehran University of Medical SciencesJournal of Biostatistics and Epidemiology2383-41962383-420X2019-03-0144Assessing Diagnostic Accuracy of Doctors Without a Gold Standard using Bayesian Networks and Kmodes Dlustering AlgorithmParisa Niloofar0Parastoo Niloofar1Mehdi Yaseri2Department of Statistics, University of Bojnord, Bojnord, Iran.Department of Epidemiology and Biostatistics,School of Public Health Tehran University of Medical Sciences, Tehran, IranDepartment of Epidemiology and Biostatistics, School of Public Health Tehran University of Medical Sciences, Tehran, Iran Background & Aim: The diagnostic accuracy of a test is the ability to discriminate accuratelybetween patients who have and do not have the target disease. A common problem in assessing thediagnostic accuracy of doctors is the unknown true disease status which in the literature is referredas “absence of a gold standard”. Methods & Material: In this article, a Naïve Bayesian network with hidden class node and a clusteringbased algorithm for categorical data named K-modes are proposed for estimating the diagnosticaccuracy of 5 physicians in diagnosing Diabetic Retinopathy. Also to assess and compare the efficiencies of these models, a simulation study with two different scenarios is conducted. Results: Simulation study indicates that for Naïve Bayesian network and the non-rare disease, say forprevalence 0.1 and 0.2, as the sample size increases so the coverage probability. But for high prevalencevalues, say 0.5, coverage probabilities are not as good as those of non-rare disease. K-modes algorithm's efficiency decreases by the increase in the number of records, but it achieves betterresults when there are a small number of records, prevalence is approximately 0.3 and sensitivitiesare high. Results of the real data set reveal that sensitivities for all physicians except one, were higher than 85% and all specificities were higher than 90%. Also the estimated prevalence happensto be 0.32. Conclusion: Through simulations and data analysis we show that this new approach based on Naïve Bayesian networks provides a useful alternative to traditional latent class modeling approaches usedin this setting. https://jbe.tums.ac.ir/index.php/jbe/article/view/226Bayesian networksCluster AnalysisDiabetic RetinopathyHumansSensitivity |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Parisa Niloofar Parastoo Niloofar Mehdi Yaseri |
spellingShingle |
Parisa Niloofar Parastoo Niloofar Mehdi Yaseri Assessing Diagnostic Accuracy of Doctors Without a Gold Standard using Bayesian Networks and Kmodes Dlustering Algorithm Journal of Biostatistics and Epidemiology Bayesian networks Cluster Analysis Diabetic Retinopathy Humans Sensitivity |
author_facet |
Parisa Niloofar Parastoo Niloofar Mehdi Yaseri |
author_sort |
Parisa Niloofar |
title |
Assessing Diagnostic Accuracy of Doctors Without a Gold Standard using Bayesian Networks and Kmodes Dlustering Algorithm |
title_short |
Assessing Diagnostic Accuracy of Doctors Without a Gold Standard using Bayesian Networks and Kmodes Dlustering Algorithm |
title_full |
Assessing Diagnostic Accuracy of Doctors Without a Gold Standard using Bayesian Networks and Kmodes Dlustering Algorithm |
title_fullStr |
Assessing Diagnostic Accuracy of Doctors Without a Gold Standard using Bayesian Networks and Kmodes Dlustering Algorithm |
title_full_unstemmed |
Assessing Diagnostic Accuracy of Doctors Without a Gold Standard using Bayesian Networks and Kmodes Dlustering Algorithm |
title_sort |
assessing diagnostic accuracy of doctors without a gold standard using bayesian networks and kmodes dlustering algorithm |
publisher |
Tehran University of Medical Sciences |
series |
Journal of Biostatistics and Epidemiology |
issn |
2383-4196 2383-420X |
publishDate |
2019-03-01 |
description |
Background & Aim: The diagnostic accuracy of a test is the ability to discriminate accuratelybetween patients who have and do not have the target disease. A common problem in assessing thediagnostic accuracy of doctors is the unknown true disease status which in the literature is referredas “absence of a gold standard”.
Methods & Material: In this article, a Naïve Bayesian network with hidden class node and a clusteringbased algorithm for categorical data named K-modes are proposed for estimating the diagnosticaccuracy of 5 physicians in diagnosing Diabetic Retinopathy. Also to assess and compare the efficiencies of these models, a simulation study with two different scenarios is conducted.
Results: Simulation study indicates that for Naïve Bayesian network and the non-rare disease, say forprevalence 0.1 and 0.2, as the sample size increases so the coverage probability. But for high prevalencevalues, say 0.5, coverage probabilities are not as good as those of non-rare disease. K-modes algorithm's efficiency decreases by the increase in the number of records, but it achieves betterresults when there are a small number of records, prevalence is approximately 0.3 and sensitivitiesare high. Results of the real data set reveal that sensitivities for all physicians except one, were higher than 85% and all specificities were higher than 90%. Also the estimated prevalence happensto be 0.32.
Conclusion: Through simulations and data analysis we show that this new approach based on Naïve Bayesian networks provides a useful alternative to traditional latent class modeling approaches usedin this setting.
|
topic |
Bayesian networks Cluster Analysis Diabetic Retinopathy Humans Sensitivity |
url |
https://jbe.tums.ac.ir/index.php/jbe/article/view/226 |
work_keys_str_mv |
AT parisaniloofar assessingdiagnosticaccuracyofdoctorswithoutagoldstandardusingbayesiannetworksandkmodesdlusteringalgorithm AT parastooniloofar assessingdiagnosticaccuracyofdoctorswithoutagoldstandardusingbayesiannetworksandkmodesdlusteringalgorithm AT mehdiyaseri assessingdiagnosticaccuracyofdoctorswithoutagoldstandardusingbayesiannetworksandkmodesdlusteringalgorithm |
_version_ |
1724399492150591488 |