In hospital intensive care units, ventilator machines are used to support breathing functions in patients who are in critical condition. When a patient recovers from such a critical status, the readiness of the patient to breathe autonomously is verified via a procedure called Spontaneous Breathing Trial (SBT). If the procedure concludes successfully, the patient is extubated, and the ventilator machine can be allocated to a new user. Performing SBT and extubation procedures on a patient who is not ready has critical consequences on the patient's health. Thus, accurately selecting the candidate is extremely important. Such a decision is carried out by monitoring the patient's status during the hours before the maneuver, considering how his/her health parameters vary in time.

The goal of this project is to analyze the clinical data collected by the ventilator machines of the intensive care unit of Padova's hospital, considering as events of interest the attempts to perform SBT and extubation procedures. Hence, the student is asked to design an algorithm that can extract new clinical biomarkers from the series of data and possibly predict the outcome of the SBT and extubation procedures.

For more information, contact Federico Mason at federico.mason@unipd.it