Predicting Severe Sepsis Using Text from the Electronic Health Record
@article{Culliton2017PredictingSS, title={Predicting Severe Sepsis Using Text from the Electronic Health Record}, author={Phil Culliton and M. Levinson and Alice Ehresman and Joshua Wherry and J. Steingrub and S. I. Gallant}, journal={ArXiv}, year={2017}, volume={abs/1711.11536} }
Employing a machine learning approach we predict, up to 24 hours prior, a diagnosis of severe sepsis. Strongly predictive models are possible that use only text reports from the Electronic Health Record (EHR), and omit structured numerical data. Unstructured text alone gives slightly better performance than structured data alone, and the combination further improves performance. We also discuss advantages of using unstructured EHR text for modeling, as compared to structured EHR data.
Tables and Topics from this paper
Tables
8 Citations
Machine learning for early prediction of circulatory failure in the intensive care unit
- Computer Science, Mathematics
- ArXiv
- 2019
- 4
- PDF
Machine learning for clinical decision support in infectious diseases: A narrative review of current applications.
- Medicine
- Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases
- 2019
- 23
Machine learning for prediction of septic shock at initial triage in emergency department.
- Medicine
- Journal of critical care
- 2019
- 5
High-performance medicine: the convergence of human and artificial intelligence
- Computer Science, Medicine
- Nature Medicine
- 2019
- 802
- PDF
The Application of Deep Learning in Cancer Prognosis Prediction
- Computer Science, Medicine
- Cancers
- 2020
- 16
- PDF
Fighting healthcare rocketing costs with value-based medicine: the case of stroke management
- Medicine
- BMC Health Services Research
- 2020
References
SHOWING 1-10 OF 12 REFERENCES
Prediction of Sepsis in the Intensive Care Unit With Minimal Electronic Health Record Data: A Machine Learning Approach
- Medicine
- JMIR medical informatics
- 2016
- 176
- PDF
Developing Predictive Models Using Electronic Medical Records: Challenges and Pitfalls
- Medicine, Computer Science
- AMIA
- 2013
- 71
- PDF
Risk Stratification of ICU Patients Using Topic Models Inferred from Unstructured Progress Notes
- Medicine, Computer Science
- AMIA
- 2012
- 79
- PDF
Creating an automated trigger for sepsis clinical decision support at emergency department triage using machine learning
- Medicine
- PloS one
- 2017
- 107
- PDF
Realizing the full potential of electronic health records: the role of natural language processing
- Medicine, Computer Science
- J. Am. Medical Informatics Assoc.
- 2011
- 45
- PDF
MIMIC-III, a freely accessible critical care database
- Medicine, Computer Science
- Scientific data
- 2016
- 2,000
- PDF
Incidence and Trends of Sepsis in US Hospitals Using Clinical vs Claims Data, 2009-2014
- Medicine
- JAMA
- 2017
- 464
- PDF
A targeted real-time early warning score (TREWScore) for septic shock
- Medicine
- Science Translational Medicine
- 2015
- 286
- PDF