Exploiting Text Data to Improve Critical Care Mortality Prediction

Auslander, Bryan; Gupta, Kalyan Moy; Floyd, Michael W.; Blisard, Sam; Aha, David W.

doi:10.1109/gcwkshps50303.2020.9367547

Cited by 1 publication

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They have achieved 0.782 AUC. In [50], the authors have also used unstructured clinical notes for mortality prediction. They have used word2vec [51] and bag-of-words as feature extraction methods, and XGBoost and Logistic Regression as machine learning models.…”

Section: Related Workmentioning

confidence: 99%

“…As shown, for TF-IDF, the best AU-ROC scores are given by the following combinations: C = 0.5 for logistic regression; num_leaves = 2, max_depth = [2,10,50], learning_rate = 0.1, n_estimators = 500, min_child_weight = [0.001, 0.1, 0.5], min_child_samples = 50 for light-GBM; n_estimator = 500, max_depth = None, max_features = ['auto', 'sqrt'], min_samples_split = 2, min_samples_leaf = 1 for random forest; learning_rate = 0.02, hidden_nodes = 200 for feed-forward neural network.…”

Section: Plos Onementioning

confidence: 99%

“…When multiple values of a certain hyperparameter help the model achieve the highest AU-ROC score, such as max_depth = [2,10,50] for light-GBM, we randomly choose the first value in the list as the optimal one.…”

Section: Plos Onementioning

confidence: 99%

See 2 more Smart Citations

Unstructured clinical notes within the 24 hours since admission predict short, mid & long-term mortality in adult ICU patients

et al. 2022

View full text Add to dashboard Cite

Mortality prediction for intensive care unit (ICU) patients is crucial for improving outcomes and efficient utilization of resources. Accessibility of electronic health records (EHR) has enabled data-driven predictive modeling using machine learning. However, very few studies rely solely on unstructured clinical notes from the EHR for mortality prediction. In this work, we propose a framework to predict short, mid, and long-term mortality in adult ICU patients using unstructured clinical notes from the MIMIC III database, natural language processing (NLP), and machine learning (ML) models. Depending on the statistical description of the patients’ length of stay, we define the short-term as 48-hour and 4-day period, the mid-term as 7-day and 10-day period, and the long-term as 15-day and 30-day period after admission. We found that by only using clinical notes within the 24 hours of admission, our framework can achieve a high area under the receiver operating characteristics (AU-ROC) score for short, mid and long-term mortality prediction tasks. The test AU-ROC scores are 0.87, 0.83, 0.83, 0.82, 0.82, and 0.82 for 48-hour, 4-day, 7-day, 10-day, 15-day, and 30-day period mortality prediction, respectively. We also provide a comparative study among three types of feature extraction techniques from NLP: frequency-based technique, fixed embedding-based technique, and dynamic embedding-based technique. Lastly, we provide an interpretation of the NLP-based predictive models using feature-importance scores.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Plos Onementioning

confidence: 99%