Handling missing data in large healthcare dataset: A case study of unknown trauma outcomes

@article{Mirkes2016HandlingMD,
  title={Handling missing data in large healthcare dataset: A case study of unknown trauma outcomes},
  author={Eugenij Moiseevich Mirkes and Timothy J. Coats and Jeremy Levesley and Alexander N Gorban},
  journal={Computers in biology and medicine},
  year={2016},
  volume={75},
  pages={
          203-16
        }
}
A Pragmatic Ensemble Strategy for Missing Values Imputation in Health Records
TLDR
A unique Ensemble Strategy for Missing Value is introduced to analyse healthcare data with considerable missing values to identify unbiased and accurate prediction statistical modelling and indicates that the proposed technique surpasses standard missing value imputation approaches as well as the approach of dropping records holding missing values in terms of accuracy.
A Hybrid Imputation Method for Multi-Pattern Missing Data: A Case Study on Type II Diabetes Diagnosis
TLDR
A four-layer model is introduced, and a hybrid imputation (HIMP) method using this model is proposed to impute multi-pattern missing data including non-random, random, and completely random patterns.
What can the randomness of missing values tell you about clinical practice in large data sets of children’s vital signs?
TLDR
The paediatric observation priority score (POPS) is a methodology to assess the acuity of children presenting to urgent and emergency care environments and creates opportunities to improve patient safety through automated alert systems and also provides mechanisms for better quality improvement and assurance.
iHealthcare: Predictive Model Analysis Concerning Big Data Applications for Interactive Healthcare Systems †
TLDR
A piece of healthcare technology which can deal with a patient's past and present medical data including symptoms of a disease, emotional data, and genetic data is designed and a prediction scheme that is performed in the cloud server to predict disease in a patient is presented.
Leveraging healthcare utilization to explore outcomes from musculoskeletal disorders: methodology for defining relevant variables from a health services data repository
TLDR
The methodology related to the use of the United States Military Health System Data Repository for longitudinal assessment of musculoskeletal clinical outcomes is discussed, as well as challenges of using this data for outcomes research are addressed.
An Interactive Healthcare System of Big Data Application with Predictive Model Analysis
TLDR
A healthcare system to deal with patients biological, and emotional condition as well as the previous health history with genetical data is designed and a prediction algorithm which is performed in cloud server to predict a patients disease is presented.
Machine Learning for the Deterioration of Patients on Hospital Wards
TLDR
The results suggest that accounting for ageand sex-related vital sign changes more can accurately detect deterioration of non-elderly patients prior to an adverse event than current methods.
Machine Learning for Clinical Outcome Prediction
TLDR
The state-of-the-art in related works covering data processing, inference, and model evaluation, in the context of outcome prediction models developed using data extracted from electronic health records are summarized.
...
1
2
3
4
...

References

SHOWING 1-10 OF 75 REFERENCES
The impact of missing trauma data on predicting massive transfusion
TLDR
Evaluating the accuracy clinical prediction models with missing data can be misleading, especially with many predictor variables and moderate levels of missingness per variable, according to the proposed sensitivity analysis.
Missing data in medical databases: Impute, delete or classify?
Bias arising from missing data in predictive models.
  • M. Gorelick
  • Medicine
    Journal of clinical epidemiology
  • 2006
A new approach to outcome prediction in trauma: A comparison with the TRISS model.
TLDR
The new model has enabled most of the cases that were excluded under the TRISSs inclusion criteria to be included, less missing data are incurred and the predictive performance was significantly better than that of the TRiss model as shown by the AROC curves.
Recent massive blood transfusion practice in England and Wales: view from a trauma registry
TLDR
The authors were unable to show that FFP and platelet use were significant predictors of survival in MBT, and it is concluded that MBT is a rare event with high mortality in UK trauma.
Different definitions of patient outcome: consequences for performance analysis in trauma.
Deaths from trauma in London—a single centre experience
TLDR
Despite current focus on death from knife and gun crime, the vast majority of trauma mortality arises from blunt aetiology and Maturation of the authors' systems of care has been associated with a drop in mortality as institutional trauma volumes increase and clinical infrastructure develops.
Purposeful variable selection and stratification to impute missing Focused Assessment with Sonography for Trauma data in trauma research
BACKGROUND The Focused Assessment with Sonography for Trauma (FAST) examination is an important variable in many retrospective trauma studies. The purpose of this study was to devise an imputation
...
1
2
3
4
5
...