Thomas Schmidt

Thomas Schmidt (2)The Maersk Mc-Kinney Moeller Institute (MMMI)
Faculty of Engineering
University of Southern Denmark





The PhD was concluded on February 10 2016 with a formal defence proceeding.

Thomas is now working as a Post.Doc at The Maersk Mc-Kinney Moeller Institute (MMMI), where he is involved in the Patient@Home project, ACQUIRE-HF which will develop a digital platform to improve the treatment and quality of life for heart failure patients.


Read about Thomas’ PhD project here:

Identification of High-risk Patients: Analysis of patient monitors with the view of developing a new model for identification of patients at risk

Almost all citizens being acutely hospitalized at Odense University Hospital are triaged at the Joint Emergency Reception (FAM). Patients are classified in five categories – red, orange, yellow, green, and blue – depending on the severity of their condition. Patients in the red category are referred directly to the hospital’s trauma room, while patients from the orange and yellow categories are referred directly for treatment and often connected to a patient monitor that continuously monitors the vital parameters of the patient. Patients from the green and blue categories are reassessed at a later point of time. The challenges of this method are situations when some patients, who were not initially assessed as high-risk patients, suddenly and unexpectedly get worse – sometimes with severe consequences.

The Vision

The vision of the project is to give healthcare professionals at Odense University Hospital new possibilities to identify FAM patients with deteriorating vital parameters suchas blood pressure, pulse, saturation (oxygen saturation), respiratory rate, GCS (Glaskow Coma Scale) and temperature. The goal is to develop novel technological models that are able to better predict and warn of potential life-threatening complications before they occur.

Analyses of data from patient monitors will help the project discover patterns and develop a new and better model for the identification of patients at risk. Subsequently, the model will be evaluated through tests of “live” data from patient monitors.

Permanent link to this article: