Predicting Severe Sepsis Using Text from the Electronic Health Record

11/30/2017

∙

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.

READ FULL TEXT

Predicting Severe Sepsis Using Text from the Electronic Health Record

Deep EHR: Chronic Disease Prediction Using Medical Notes

Identification of Predictive Sub-Phenotypes of Acute Kidney Injury using Structured and Unstructured Electronic Health Record Data with Memory Networks

A Topic Modeling Approach to Classifying Open Street Map Health Clinics and Schools in Sub-Saharan Africa

Using Machine Learning to Fuse Verbal Autopsy Narratives and Binary Features in the Analysis of Deaths from Hyperglycaemia

Navigating the Data Lake with Datamaran: Automatically Extracting Structure from Log Datasets

Prediction Using Note Text: Synthetic Feature Creation with word2vec

Prompt-based Extraction of Social Determinants of Health Using Few-shot Learning

Predicting Severe Sepsis Using Text from the Electronic Health Record

Related Research

Deep EHR: Chronic Disease Prediction Using Medical Notes

Identification of Predictive Sub-Phenotypes of Acute Kidney Injury using Structured and Unstructured Electronic Health Record Data with Memory Networks

A Topic Modeling Approach to Classifying Open Street Map Health Clinics and Schools in Sub-Saharan Africa

Using Machine Learning to Fuse Verbal Autopsy Narratives and Binary Features in the Analysis of Deaths from Hyperglycaemia

Navigating the Data Lake with Datamaran: Automatically Extracting Structure from Log Datasets

Prediction Using Note Text: Synthetic Feature Creation with word2vec

Prompt-based Extraction of Social Determinants of Health Using Few-shot Learning