Using methods from the data-mining and machine-learning literature for disease classification and prediction: a case study examining classification of heart failure subtypes
Journal of Clinical EpidemiologyPublished 4 February 2013
Peter C. Austin, Jack V. Tu, Jennifer E. Ho, Daniel Levy, Douglas S. Lee
Citations342
SJR quartileQ1
SJR score3.15
SNIP2.66
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
It is found that modern, flexible tree-based methods from the data-mining literature offer substantial improvement in prediction and classification of HF subtype compared with conventional classification and regression trees, but these methods do not offer substantial improvements over logistic regression for predicting the presence of HFPEF.
Abstract
The use of tree-based methods offers superior performance over conventional classification and regression trees for predicting and classifying HF subtypes in a population-based sample of patients from Ontario, Canada. However, these methods do not offer substantial improvements over logistic regression for predicting the presence of HFPEF.
Keywords
Computer ScienceHealth Professions
Machine LearningRandom Forests
126,192 Citations2001Leo Breiman
Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the forest, and are also applicable to regression.
Springer series in statisticsThe Elements of Statistical Learning
24,344 Citations2001Trevor Hastie, J. Friedman +1 more
BiometricsClassification and Regression Trees.
23,841 Citations1984Alexander Gordon, Leo Breiman +3 more
The Elements of Statistical Learning: Data Mining, Inference, and Prediction
19,345 Citations2013Trevor Hastie, Robert Tibshirani +1 more
Classification and Regression by randomForest
18,390 Citations2007Andy Liaw, Matthew C. Wiener
random forests are proposed, which add an additional layer of randomness to bagging and are robust against overfitting, and the randomForest package provides an R interface to the Fortran programs by Breiman and Cutler.
Journal of the American College of Cardiology2013 ACCF/AHA Guideline for the Management of Heart Failure
12,625 Citations2013Clyde W. Yancy, Mariell Jessup +22 more
The 2017-18 FACC-FAHA curriculum vitae will be presented in June, with a focus on teaching, research, and awareness of adolescent and young people’s experiences with FACC.
Experiments with a new boosting algorithm
7,585 Citations1996Yoav Freund, Robert E. Schapire
The Annals of StatisticsAdditive logistic regression: a statistical view of boosting (With discussion and a rejoinder by the authors)
6,842 Citations2000Jerome H. Friedman, Trevor Hastie +1 more
This work shows that this seemingly mysterious phenomenon of boosting can be understood in terms of well-known statistical principles, namely additive modeling and maximum likelihood, and develops more direct approximations and shows that they exhibit nearly identical results to boosting.
An empirical comparison of supervised learning algorithms
2,714 Citations2006Rich Caruana, Alexandru Niculescu-Mizil
A large-scale empirical comparison between ten supervised learning methods: SVMs, neural nets, logistic regression, naive bayes, memory-based learning, random forests, decision trees, bagged trees, boosted trees, and boosted stumps is presented.
Wiley series in probability and statisticsStatistical Methods in Diagnostic Medicine
1,789 Citations2011Xiao‐Hua Zhou, Nancy A. Obuchowski +3 more
This book discusses the design of Diagnostic Accuracy Studies, the construction of a Smooth ROC Curve, and how to select a Sampling Plan for Readers based on Sensitivity and Specificity.
Psychological MethodsPropensity Score Estimation With Boosted Regression for Evaluating Causal Effects in Observational Studies.
1,318 Citations2004Daniel F. McCaffrey, Greg Ridgeway +1 more
Propensity score weights estimated using boosting eliminate most pretreatment group differences and substantially alter the apparent relative effects of adolescent substance abuse treatment.
Artificial Intelligence in MedicinePredicting breast cancer survivability: a comparison of three data mining methods
1,216 Citations2004Dursun Delen, Glenn Walker +1 more
The comparative study of multiple prediction models for breast cancer survivability using a large dataset along with a 10-fold cross-validation provided us with an insight into the relative prediction ability of different data mining methods.
Journal of the American College of CardiologyHeart failure with preserved left ventricular systolic function
937 Citations2004Karen Hogg, Karl Swedberg +1 more
Though much has recently been learned about the syndrome of HF-PSF, many questions remain to be answered, not least how it should be treated.
Annals of Behavioral MedicineClassification and regression tree analysis in public health: Methodological review and comparison with logistic regression
874 Citations2003Stephenie C. Lemon, Jason Roy +3 more
C&RT is a promising research tool for the identification of at-risk populations in public health research and outreach and is compared to a common approach to logistic regression, the stepwise selection procedure.
CirculationRelation of Disease Pathogenesis and Risk Factors to Heart Failure With Preserved or Reduced Ejection Fraction
675 Citations2009Douglas S. Lee, Philimon Gona +6 more
It is suggested thatheart failure with reduced left ventricular systolic function and heart failure with preserved left Ventricular syStolic function are partially distinct entities, with potentially different approaches to early detection and prevention.
Tree-Based Models
559 Citations2017Linda A. Clark, Daryl Pregibon
An empirical evaluation of supervised learning in high dimensions
531 Citations2008Rich Caruana, Nikos Karampatziakis +1 more
To the surprise, the method that performs consistently well across all dimensions is random forests, followed by neural nets, boosted trees, and SVMs, and the effect of increasing dimensionality on the performance of the learning algorithms changes.
Classification and Regression Trees
506 Citations2022Elinor Jones, Simon Harden +1 more
BMC Research NotesData mining methods in the prediction of Dementia: A real-data comparison of the accuracy, sensitivity and specificity of linear discriminant analysis, logistic regression, neural networks, support vector machines, classification trees and random forests
445 Citations2011João Marôco, Dina Silva +4 more
When taking into account sensitivity, specificity and overall classification accuracy Random Forests and Linear Discriminant analysis rank first among all the classifiers tested in prediction of dementia using several neuropsychological tests.
Journal of the American College of CardiologyGender, age, and heart failure with preserved left ventricular systolic function
423 Citations2003Frederick A. Masoudi, Edward P. Havranek +5 more
In elderly patients hospitalized with HF, preserved systolic function is primarily a condition of women, independent of important demographic and clinical characteristics.
Medical CarePrediction Modeling Using EHR Data
382 Citations2010Jionglin Wu, Jason Roy +1 more
Heart failure was predicted more than 6 months before clinical diagnosis, with AUC of about 0.76, using logistic regression and Boosting, and SVM had the poorest performance, possibly because of imbalanced data.
JAMAEffectiveness of Public Report Cards for Improving the Quality of Cardiac Care
258 Citations2009Jack V. Tu, Linda R. Donovan +5 more
Statistics in MedicineA comparison of regression trees, logistic regression, generalized additive models, and multivariate adaptive regression splines for predicting AMI mortality
155 Citations2006Peter C. Austin
It is shown that regression trees do not perform as well as logistic regression for predicting mortality following AMI, however, the logistic regressors had performance comparable to that of more flexible, data‐driven models such as GAMs and MARS.
European Heart JournalDiscriminating clinical features of heart failure with preserved vs. reduced ejection fraction in the community
153 Citations2012Jennifer E. Ho, Philimon Gona +8 more
Multiple clinical characteristics at the time of initial HF presentation differed in participants with HFPEF vs. HFREF, and while CHD was clearly associated with a lower EF, overlapping characteristics were observed in the middle of the left ventricular EF range spectrum.
Biometrical JournalRegression trees for predicting mortality in patients with cardiovascular disease: What improvement is achieved by using ensemble‐based methods?
78 Citations2012Peter C. Austin, Douglas S. Lee +2 more
It is concluded that ensemble methods from the data mining and machine learning literature increase the predictive performance of regression trees, but may not lead to clear advantages over conventional logistic regression models for predicting short‐term mortality in population‐based samples of subjects with cardiovascular disease.
Journal of the American College of CardiologyCholesterol emboli after invasive cardiac procedures**Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
55 Citations2003Thomas M. Bashore, Thomas Gehrig
The term “atheroemboli” is used to refer to the dislodgement of vascular plaque material that contains cholesterol crystals plus red blood cells and fibrin that can occlude major systemic vessels.
Journal of Clinical EpidemiologyRecursive Partitioning for the Identification of Disease Risk Subgroups: A Case-Control Study of Subarachnoid Hemorrhage
54 Citations1998Lorene M. Nelson, D. Blöch +2 more
The utility of recursive partitioning as a supplement to logistic regression for the multivariable analysis of data from case-control studies is examined, demonstrating the technique using data from women enrolled in a population-based study of subarachnoid hemorrhage.
Advances in Dental ResearchDental Data Mining: Potential Pitfalls and Practical Issues
27 Citations2003Stuart A. Gansky
The performance of logistic regression with KDD methods of CART and ANN in analyzing data from the Rochester caries study is compared to show that ANN and CART models can perform better than classic regression models.
PubMedDifferentiation of benign and malignant breast tumors by logistic regression and a classification tree using Doppler flow signals.
25 Citations1998Willi Sauerbrei, H. Madjar +1 more
Methods of Information in MedicineDifferentiation of Benign and Malignant Breast Tumors by Logistic Regression and a Classification Tree using Doppler flow signals
20 Citations1998H. Madjar, H Prömpeler +1 more
The results indicate that the estimates of sensitivity and specificity are probably close to realistic values for a clinical setting, and the magnitude of the overoptimism of the diagnostic rules with resampling methods is estimated.
PubMedBoosted classification trees result in minor to modest improvement in the accuracy in classifying cardiovascular outcomes compared to conventional classification trees.
13 Citations2011Peter C. Austin, Douglas S. Lee
Gains in predictive accuracy for predicting cardiovascular outcomes were less impressive than gains in performance observed in the data mining literature and boosted classification trees had high specificity, but low sensitivity.