Missing Data Imputation for Individualised CVD Diagnostic and Treatment

Venkatraman, Sitalakshmi; Yatsko, Andrew; Stranieri, Andrew; Jelinek, Herbert F.

doi:10.22489/cinc.2016.100-179

Cited by 7 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These were substituted as previously reported. [28,29] In this evaluation DM exemplifies the class attribute. The prevalence of DM is 18% in the featured population.…”

Section: Discussionmentioning

confidence: 99%

“…From the domain knowledge [28,29] the final feature-set in Table 1 contains features regarded to be fair predictors of DM. At the same time, some of their 'duplicates' were removed.…”

Section: Discussionmentioning

confidence: 99%

“…This version is based on the body mass index (BMI) and prevails because the other one is not corrected for use of medication affecting the cholesterol components replacing BMI in the risk formula. [29] The risk variables are anticipated to top the listing, though, as there is a correction in place for DM. This inconsistency, despite being slight, compromises either version of the risk as a DM predictor variable, but using them in the probe capacity is appropriate here.…”

Section: Discussionmentioning

confidence: 99%

“…[15] The data has to be in full supply for the algorithm to work. Methods of data completion in situations when some of it is missing were previously developed [8,28,29] and some indeed applied to the test data in the current work.…”

Section: Discussionmentioning

confidence: 99%

See 3 more Smart Citations

Adaptations of Relief for continuous domains of bioinformatics

Yatsko

2019

AIR

Self Cite

View full text Add to dashboard Cite

Relief occupies a niche among feature selection methods for data classification. Filters are faster, wrappers are much slower. Relief is feature-set-aware, same as wrappers. However, it is thought being able to deselect only irrelevant, but not redundant features, same as filters. Iterative Reliefs seek to increase the separation margin between classes in the anisotropic space defined by weighted features. Reliefs for continuous domains are much less developed than for categorical domains. The paper discusses a number of adaptations for continuous spaces with Euclidean or Manhattan metric. The ability of Relief to detect redundant features is demonstrated. A dramatic reduction of the feature-set is achieved in a health diagnostics problem.

show abstract

“…These were substituted as previously reported. [28,29] In this evaluation DM exemplifies the class attribute. The prevalence of DM is 18% in the featured population.…”

Section: Discussionmentioning

confidence: 99%

“…From the domain knowledge [28,29] the final feature-set in Table 1 contains features regarded to be fair predictors of DM. At the same time, some of their 'duplicates' were removed.…”

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Adaptations of Relief for continuous domains of bioinformatics

Yatsko

2019

AIR

Self Cite

View full text Add to dashboard Cite

show abstract

“…Kuppusamy and Paramasivam (2016) introduced missing values randomly in the nursery dataset and predicted these missing values with MissForest algorithm. Venkatraman et al (2016) performed prediction of cardiovascular disease on DiabHealth dataset. Missing data were filled by using the mean method for continuous features and the mode method for categorical features.…”

Section: Related Workmentioning

confidence: 99%

HIOC: a hybrid imputation method to predict missing values in medical datasets

Rani

Kumar

Jain

2021

IJICC

View full text Add to dashboard Cite

PurposeDecision support systems developed using machine learning classifiers have become a valuable tool in predicting various diseases. However, the performance of these systems is adversely affected by the missing values in medical datasets. Imputation methods are used to predict these missing values. In this paper, a new imputation method called hybrid imputation optimized by the classifier (HIOC) is proposed to predict missing values efficiently.Design/methodology/approachThe proposed HIOC is developed by using a classifier to combine multivariate imputation by chained equations (MICE), K nearest neighbor (KNN), mean and mode imputation methods in an optimum way. Performance of HIOC has been compared to MICE, KNN, and mean and mode methods. Four classifiers support vector machine (SVM), naive Bayes (NB), random forest (RF) and decision tree (DT) have been used to evaluate the performance of imputation methods.FindingsThe results show that HIOC performed efficiently even with a high rate of missing values. It had reduced root mean square error (RMSE) up to 17.32% in the heart disease dataset and 34.73% in the breast cancer dataset. Correct prediction of missing values improved the accuracy of the classifiers in predicting diseases. It increased classification accuracy up to 18.61% in the heart disease dataset and 6.20% in the breast cancer dataset.Originality/valueThe proposed HIOC is a new hybrid imputation method that can efficiently predict missing values in any medical dataset.

show abstract