Predicting In-Hospital Mortality of ICU Patients: The PhysioNet/Computing in Cardiology Challenge 2012

By Ikaro Silva, George B. Moody, Daniel J. Scott, Leo Anthony Celi, R. Mark. 2012

436 citations recorded by the source corpus.

Acuity scores, such as APACHE, SAPS, MPM, and SOFA, are widely used to account for population differences in studies aiming to compare how medications, care guidelines, surgery, and other interventions impact mortality in Intensive Care Unit (ICU) patients. By contrast, the focus of the PhysioNet/CinC Challenge 2012 is to develop methods for patient-specific prediction of in-hospital mortality. The data used for the challenge consisted of 5 general descriptors and 36 time series (measurements of vital signs and laboratory results) from the first 48 hours of the first available ICU stay of 12,000 adult patients from the MIMIC II database. The challenge was organized as two events: event 1 measured performance of a binary classifier, and event 2 measured performance of a risk estimator. The score of event 1 was the lower of sensitivity and positive predictive value. The score for event 2 was a range-normalized Hosmer-Lemeshow statistic. A baseline algorithm (using SAPS-1) obtained event 1 and 2 scores of 0.3125 and 68.58 respectively. Most participants submitted entries that outperformed the baseline algorithm. The top final scores for events 1 and 2 were 0.5353 and 17.88 respectively.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Learning Credible Models (2017, methodology): “We use the ICU data provided in the PhysioNet Challenge 2012 [27] to train our model.”
Recurrent Neural Networks for Multivariate Time Series with Missing Values (2016, methodology): “Physionet Challenge 2012 dataset (PhysioNet) This dataset, from PhysioNet Challenge 2012 (Silva et al., 2012), is a publicly available collection of multivariate clinical time series from 8000 intensive care unit (ICU) records.”
Latent ODEs for Irregularly-Sampled Time Series (2019, methodology): “We evaluated our model on the PhysioNet Challenge 2012 dataset [Silva et al., 2012], which contains 8000 time series, each containing measurements from the first 48 hours of a different patient’s admission to ICU.”
Deep Computational Phenotyping (2015, methodology): “The first data set comes from PhysioNet Challenge 2012 website [30] which is a publicly available(1) collection of multivariate clinical time series from 8000 ICU units.”
BRITS: Bidirectional Recurrent Imputation for Time Series (2018, methodology): “2 Health-care Data We evaluate our models on health-care data in PhysioNet Challenge 2012 [27], which consists of 4000 multivariate clinical time series from intensive care unit (ICU).”
Learning Differential Equations that are Easy to Solve (2020, methodology): “As in Rubanova et al. (2019), we use the Latent ODE architecture for modelling trajectories of ICU patients using the PhysioNet Challenge 2012 dataset (Silva et al., 2012).”
Latent Ordinary Differential Equations for Irregularly-Sampled Time Series (2019, methodology): “We evaluate our approach on the PhysioNet Challenge 2012 (Silva et al., 2012) dataset of 8000 patients from Intensive Care Unit (ICU).”
GP-VAE: Deep Probabilistic Time Series Imputation (2019, methodology): “We also applied our model to the data set from the 2012 Physionet Challenge [43].”
SAITS: Self-Attention-based Imputation for Time Series (2022, methodology): “…failure of collection sensors, communication error, and unexpected malfunction, missing values are common to see in time series from the real-world environment, for example, patient health measurement [1], air-quality monitoring [2], telecommunication networking [3], and educational systems [4].”
Machine Learning and Decision Support in Critical Care (2016, background): “The 2007 Computational Medicine Challenge provided a corpus of de-identified radiology reports and gave participants the task of assigning two codes from a set of 45 ICD-9 codes [35].”