• Publications
  • Influence
MIMIC-III, a freely accessible critical care database
MIMIC-III (‘Medical Information Mart for Intensive Care’) is a large, single-center database comprising information relating to patients admitted to critical care units at a large tertiary care
The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care
A reinforcement learning agent, the AI Clinician, can assist physicians by providing individualized and clinically interpretable treatment decisions to improve patient outcomes by extracting implicit knowledge from an amount of patient data that exceeds by many-fold the life-time experience of human clinicians.
The eICU Collaborative Research Database, a freely available multi-center database for critical care research
The e ICU Collaborative Research Database is described, a multi-center intensive care unit (ICU) database with high granularity data for over 200,000 admissions to ICUs monitored by eICU Programs across the United States.
Predicting in-hospital mortality of ICU patients: The PhysioNet/Computing in cardiology challenge 2012
The focus of the PhysioNet/CinC Challenge 2012 is to develop methods for patient-specific prediction of in-hospital mortality by measuring performance of a binary classifier and a risk estimator.
Human cerebrovascular and ventilatory CO2 reactivity to end‐tidal, arterial and internal jugular vein PCO2
The hypercapnic and hypocapnic MCAv‐CO2 reactivity was higher (∼97% and ∼24%, respectively) when expressed with P  jv,CO 2 than P’s ’a, CO’2 (P < 0.05), indicating that a reduced reactivity results in less central CO2 washout and greater ventilatory stimulus.
Improving the quality of ECGs collected using mobile phones: The PhysioNet/Computing in Cardiology Challenge 2011
The aim of the PhysioNet/Computing in Cardiology Challenge 2011 was to develop an efficient algorithm able to run within a mobile phone, that can provide useful feedback in the process of acquiring a
Early intervention with erythropoietin does not affect the outcome of acute kidney injury (the EARLYARF trial).
The study showed that a composite of these two biomarkers was insufficient for risk stratification in a patient population with a heterogeneous onset of injury, and early intervention with high-dose erythropoietin was safe but did not alter the outcome.
ICU admission characteristics and mortality rates among elderly and very elderly patients
The proportion of elderly patients from the total ICU population is high, and with advancing age, the proportion of various preexisting comorbidities and the primary reason for ICU admission change should be regarded as a significant independent risk factor for mortality, especially for ICu patients older than 75.
Mechanical power of ventilation is associated with mortality in critically ill patients: an analysis of patients in two observational cohorts
Mechanical power of ventilation is independently associated with higher in-hospital mortality and several other outcomes in ICU patients receiving invasive ventilation for at least 48 h.
Deep Reinforcement Learning for Sepsis Treatment
This work proposes an approach to deduce treatment policies for septic patients by using continuous state-space models and deep reinforcement learning, which learns clinically interpretable treatment policies, similar in important aspects to the treatment policies of physicians.