Using permutations to assess confounding in machine learning applications for digital health.
Explore this paper's citation graph
Summary
Novel permutation based statistical methods are developed to detect and quantify the influence of observed confounders, and estimate the unconfounded performance of the learner.
- Type
- preprint
- Published
- 2018-11-29
- Cited by
- 6
- References
- 23
- Access
- Open access
- OpenAlex
- https://openalex.org/W2901986740
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:88523333
Keywords
Confounding, Computer science, Generalizability theory, Weighting, Machine learning
References
- Multivariate Bernoulli distribution
- The area above the ordinal dominance graph and the area below the receiver operating characteristic graph
- Improving propensity score weighting using machine learning
- Consistency and Unbiasedness of Certain Nonparametric Tests
- Areas beneath the relative operating characteristics (ROC) and relative operating levels (ROL) curves: Statistical significance and interpretation
- Sex-specific differences in gait patterns of healthy older adults: Results from the Baltimore Longitudinal Study of Aging
- Permutation Tests: A Practical Guide to Resampling Methods for Testing Hypotheses
- The central role of the propensity score in observational studies for causal effects
- ccSVM: correcting Support Vector Machines for confounding factors in biological data classification
- Multivariate Pattern Analysis and Confounding in Neuroimaging
- The mPower study, Parkinson disease mobile data collected using ResearchKit
- Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach.
- Smartphones as new tools in the management and understanding of Parkinson’s disease
- Robust Text Classification in the Presence of Confounding Bias
- Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs.
- Predictive modelling using neuroimaging data in the presence of confounds
- Dermatologist–level classification of skin cancer with deep neural networks
- Deep Learning Algorithms for Detection of Lymph Node Metastases From Breast Cancer: Helping Artificial Intelligence Be Seen.
- Approximation Theorems of Mathematical Statistics
- A Class of Statistics with Asymptotically Normal Distribution
Cited by
- Measuring the effects of confounders in medical supervised classification problems: the Confounding Index (CI)
- Detecting the impact of subject characteristics on machine learning-based diagnostic applications
- Causality-based tests to detect the influence of confounders on mobile health diagnostic applications: a comparison with restricted permutations
- Towards biomarkers for outcomes after pancreatic ductal adenocarcinoma and ischaemic stroke, with focus on (co)-morbidity and ageing/cellular senescence (SASKit): protocol for a prospective cohort study
- Smartphone-based digital biomarkers for Parkinson’s disease in a remotely-administered setting
- Confound-leakage: confound removal in machine learning leads to leakage
- Smartphone-based digital biomarkers for Parkinson’s disease in a remotely-administered setting
Related papers
- A Permutation Approach to Assess Confounding in Machine Learning Applications for Digital Health
- Causality-based tests to detect the influence of confounders on mobile health diagnostic applications: a comparison with restricted permutations
- Measured and Accounted-for Confounding in Pharmacoepidemiologic Studies: Some Thoughts for Practitioners
- Opportunities for minimization of confounding in observational research
- Variable Selection for Confounding Adjustment in High-dimensional Covariate Spaces When Analyzing Healthcare Databases
- Assessing the impact of unmeasured confounders for credible and reliable real‐world evidence
- Addressing unmeasured confounding in comparative observational research