Dataset Information

Inclusion of Unstructured Clinical Text Improves Early Prediction of Death or Prolonged ICU Stay.

ABSTRACT: OBJECTIVES:Early prediction of undesired outcomes among newly hospitalized patients could improve patient triage and prompt conversations about patients' goals of care. We evaluated the performance of logistic regression, gradient boosting machine, random forest, and elastic net regression models, with and without unstructured clinical text data, to predict a binary composite outcome of in-hospital death or ICU length of stay greater than or equal to 7 days using data from the first 48 hours of hospitalization. DESIGN:Retrospective cohort study with split sampling for model training and testing. SETTING:A single urban academic hospital. PATIENTS:All hospitalized patients who required ICU care at the Beth Israel Deaconess Medical Center in Boston, MA, from 2001 to 2012. INTERVENTIONS:None. MEASUREMENTS AND MAIN RESULTS:Among eligible 25,947 hospital admissions, we observed 5,504 (21.2%) in which patients died or had ICU length of stay greater than or equal to 7 days. The gradient boosting machine model had the highest discrimination without (area under the receiver operating characteristic curve, 0.83; 95% CI, 0.81-0.84) and with (area under the receiver operating characteristic curve, 0.89; 95% CI, 0.88-0.90) text-derived variables. Both gradient boosting machines and random forests outperformed logistic regression without text data (p < 0.001), whereas all models outperformed logistic regression with text data (p < 0.02). The inclusion of text data increased the discrimination of all four model types (p < 0.001). Among those models using text data, the increasing presence of terms "intubated" and "poor prognosis" were positively associated with mortality and ICU length of stay, whereas the term "extubated" was inversely associated with them. CONCLUSIONS:Variables extracted from unstructured clinical text from the first 48 hours of hospital admission using natural language processing techniques significantly improved the abilities of logistic regression and other machine learning models to predict which patients died or had long ICU stays. Learning health systems may adapt such models using open-source approaches to capture local variation in care patterns.

SUBMITTER: Weissman GE

PROVIDER: S-EPMC6005735 | biostudies-literature | 2018 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Inclusion of Unstructured Clinical Text Improves Early Prediction of Death or Prolonged ICU Stay.

Weissman Gary E GE Hubbard Rebecca A RA Ungar Lyle H LH Harhay Michael O MO Greene Casey S CS Himes Blanca E BE Halpern Scott D SD

Critical care medicine 20180701 7

<h4>Objectives</h4>Early prediction of undesired outcomes among newly hospitalized patients could improve patient triage and prompt conversations about patients' goals of care. We evaluated the performance of logistic regression, gradient boosting machine, random forest, and elastic net regression models, with and without unstructured clinical text data, to predict a binary composite outcome of in-hospital death or ICU length of stay greater than or equal to 7 days using data from the first 48 h ...[more]

PMID: 29629986

Similar Datasets

Project description:Machine learning can predict outcomes and determine variables contributing to precise prediction, and can thus classify patients with different risk factors of outcomes. This study aimed to investigate the predictive accuracy for mortality and length of stay in intensive care unit (ICU) patients using machine learning, and to identify the variables contributing to the precise prediction or classification of patients. Patients (n = 12,747) admitted to the ICU at Chiba University Hospital were randomly assigned to the training and test cohorts. After learning using the variables on admission in the training cohort, the area under the curve (AUC) was analyzed in the test cohort to evaluate the predictive accuracy of the supervised machine learning classifiers, including random forest (RF) for outcomes (primary outcome, mortality; secondary outcome, length of ICU stay). The rank of the variables that contributed to the machine learning prediction was confirmed, and cluster analysis of the patients with risk factors of mortality was performed to identify the important variables associated with patient outcomes. Machine learning using RF revealed a high predictive value for mortality, with an AUC of 0.945 (95% confidence interval [CI] 0.922-0.977). In addition, RF showed high predictive value for short and long ICU stays, with AUCs of 0.881 (95% CI 0.876-0.908) and 0.889 (95% CI 0.849-0.936), respectively. Lactate dehydrogenase (LDH) was identified as a variable contributing to the precise prediction in machine learning for both mortality and length of ICU stay. LDH was also identified as a contributing variable to classify patients into sub-populations based on different risk factors of mortality. The machine learning algorithm could predict mortality and length of stay in ICU patients with high accuracy. LDH was identified as a contributing variable in mortality and length of ICU stay prediction and could be used to classify patients based on mortality risk.

Project description:Unstructured clinical narratives are continuously being recorded as part of delivery of care in electronic health records, and dedicated tagging staff spend considerable effort manually assigning clinical codes for billing purposes. Despite these efforts, however, label availability and accuracy are both suboptimal. In this retrospective study, we aimed to automate the assignment of top-level International Classification of Diseases version 9 (ICD-9) codes to clinical records from human and veterinary data stores using minimal manual labor and feature curation. Automating top-level annotations could in turn enable rapid cohort identification, especially in a veterinary setting. To this end, we trained long short-term memory (LSTM) recurrent neural networks (RNNs) on 52,722 human and 89,591 veterinary records. We investigated the accuracy of both separate-domain and combined-domain models and probed model portability. We established relevant baseline classification performances by training Decision Trees (DT) and Random Forests (RF). We also investigated whether transforming the data using MetaMap Lite, a clinical natural language processing tool, affected classification performance. We showed that the LSTM-RNNs accurately classify veterinary and human text narratives into top-level categories with an average weighted macro F1 score of 0.74 and 0.68 respectively. In the "neoplasia" category, the model trained on veterinary data had a high validation accuracy in veterinary data and moderate accuracy in human data, with F1 scores of 0.91 and 0.70 respectively. Our LSTM method scored slightly higher than that of the DT and RF models. The use of LSTM-RNN models represents a scalable structure that could prove useful in cohort identification for comparative oncology studies. Digitization of human and veterinary health information will continue to be a reality, particularly in the form of unstructured narratives. Our approach is a step forward for these two domains to learn from and inform one another.

Project description:BackgroundPeripheral perfusion index (PPI) is an indicator reflecting perfusion. Patients undergoing long time surgeries are more prone to hypoperfusion and increased lactate. Few studies focusing on investigating the association between PPI and surgical patients' prognoses. We performed this study to find it out.MethodsFrom January 2019 to September 2019, we retrospected all surgical patients who were transferred to ICU, Xinyang Central hospital, Henan province, China. Inclusive criteria: age???18?years old; surgical length???120?min. Exclusive criteria: died in ICU; discharging against medical advice; existing diseases affecting blood flow of upper limbs, for example, vascular thrombus in arms; severe liver dysfunction. We defined "prolonged ICU stay" as patients with their length of ICU stay longer than 48?h. According to the definition, patients were divided into two groups: "prolonged group" (PG) and "non-prolong group" (nPG). Baseline characteristics, surgical and therapeutic information, ICU LOS, SOFA and APACHE II were collected. Besides we gathered data of following parameters at 3 time points (T0: ICU admission; T1: 6?h after admission; T2: 12?h after admission): mean artery pressure (MAP), lactate, heart rate (HR), PPI and body temperature. Data were compared between the 2 groups. Multivariable binary logistic regression and ROC (receiver operating characteristic) curves were performed to find the association between perfusion indictors and ICU LOS.ResultsEventually, 168 patients were included, 65 in PG and 103 in nPG. Compared to nPG, patients in PG had higher blood lactate and lower PPI. PPI showed significant difference between two groups earlier than lactate (T0 vs T1). The value of PPI at two time points was lower in PG than nPG(T0: 1.09?±?0.33 vs 1.41?±?0.45, p?= 0.001; T1: 1.08?±?0.37 vs 1.49?±?0.41, p?< 0.001). Increased lactateT1(OR 3.216; 95% CI 1.253-8.254, P?= 0.015) and decreased PPIT1 (OR 0.070; 95% CI 0.016-0.307, P <?0.001) were independently associated with prolonged ICU stay. The area under ROC of the PPIT1 for predicting ICU stay>?48?h was 0.772, and the cutoff value for PPIT1 was 1.35, with 83.3% sensitivity and 73.8% specificity.ConclusionsPPI and blood lactate at T1(6 h after ICU admission) are associated with ICU LOS in surgical patient. Compared to lactate, PPI indicates hypoperfusion earlier and more accurate in predicting prolonged ICU stay.

Dataset Information

Inclusion of Unstructured Clinical Text Improves Early Prediction of Death or Prolonged ICU Stay.

Publications

Inclusion of Unstructured Clinical Text Improves Early Prediction of Death or Prolonged ICU Stay.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets