Dynamic Prediction of ICU Mortality Risk Using Domain Adaptation

Abstract

Early recognition of risky trajectories during an Intensive Care Unit (ICU) stay is one of the key steps towards improving patient survival. Learning trajectories from physiological signals continuously measured during an ICU stay requires learning time-series features that are robust and discriminative across diverse patient populations. Patients within different ICU populations (referred here as domains) vary by age, conditions and interventions. Thus, mortality prediction models using patient data from a particular ICU population may perform suboptimally in other populations because the features used to train such models have different distributions across the groups. In this paper, we explore domain adaptation strategies in order to learn mortality prediction models that extract and transfer complex temporal features from multivariate time-series ICU data. Features are extracted in a way that the state of the patient in a certain time depends on the previous state. This enables dynamic predictions and creates a mortality risk space that describes the risk of a patient at a particular time. Experiments based on cross-ICU populations reveals that our model outperforms all considered baselines. Gains in terms of AUC range from 4% to 8% for early predictions when compared with a recent state-of-the-art representative for ICU mortality prediction. In particular, models for the Cardiac ICU population achieve AUC numbers as high as 0.88, showing excellent clinical utility for early mortality prediction. Finally, we present an explanation of factors contributing to the possible ICU outcomes, so that our models can be used to complement clinical reasoning.

Intensive Care Unit Mortality Prediction: An Improved Patient-Specific Stacking Ensemble Model

“…The irregular sampling creates a challenge for extracting data and reflects the large amount of missing data [21]. Highly imbalanced data also negatively impact our ability to obtain satisfactory results as it may lead to classifier bias in favor of the majority class [22]. All these factors impact single-classifier models differently.…”

Continuous and automatic mortality risk prediction using vital signs in the intensive care unit: a hybrid neural network approach

“…The limitations of existing scoring systems have lead to a rise in researchers exploring machine learning techniques for mortality prediction [13][14][15][16][17][18][19][20] , as well as the related issues of predicting the onset of various intervention methods 21,22 detecting the risk of sepsis [23][24][25][26] and other clinical deterioration events 27,28 . Machine learning approaches have the advantage of being relatively easy to continuously update and recalibrate, with algorithms OPEN 1 College of Science and Engineering, James Cook University, Townsville 4811, Australia.…”

“…AIMS is capable of continuously-updating prediction of the risk of mortality within 3-day, 7-day, and 14-day windows. Much of the previous literature focuses on predicting mortality events within the entire stay [15][16][17][18] , however the average length of stay in ICU in America is only 3.8 days 4 . Our analysis of the patients from the MIMIC-III database who met our selection criteria revealed that 65% of patients stayed in ICU for ≤ 3 days, 87% for ≤ 7 days, and 95% for ≤ 14 days.…”

Feature Engineering for ICU Mortality Prediction Based on Hourly to Bi-Hourly Measurements

“…Most of the ICU mortality prediction studies focus on developing powerful mortality prediction models [3][4][5][6][7][8][9][10][11][12][13][14][15][16] in which the higher priority is to provide an accurate label or score about the admitted patients' status. One drawback of such an objective is paying less attention to features' simplicity and interpretability, which is the case with deep learning approaches [7][8][9][10][11][12][13]. The key approach in these studies is black-box modelling focusing mainly on predictive model error performance, regardless the interpretability of the features.…”

“…Hence, the useful information that can be provided to the medical staff is strictly the prediction output. Moreover, a considerable number of relevant studies focus on investigating the continuously recorded vital signs of ICU patients in order to predict the mortality-risk of those patients [3][4][5][6][7][8][9][10][11][12][13][14][15][16]. A frequently used database in these studies is the medical information mart for intensive care (MIMIC) in its three releases (MIMIC, MIMIC II and III) with different versions [17,18].…”