AI-based clinical decision-making at intensive care units




Can we develop AI-based decision making solutions that support the surgical team to make faster and better decisions during a major surgery? Can we predict deterioration in hospitalized patients at intensive care units (ICUs) using deep learning? The Quantitative Healthcare Analysis (qurAI) group is seeking a PhD candidate to address these questions.

About vacancy

Intensive care units (ICUs), such as neonatal ICUs or surgical ICUs, are data-intensive environments, where intensivists must cope with a relentless flow of information to make fast and accurate decision. This information, ranging from vital signs, waveforms, demographics, test results,… needs to be combined and analyzed to make fast and accurate decisions on which patients’ life depends.  In this project we will investigate if AI models can analyze this data and provide decision-making support.

What are you going to do?

As part of your PhD, you will develop new deep learning solutions for the analysis of different waveforms, such as blood pressure waves or thromboelastograms, to diagnose the current status of the patient, predict medical complications or conditions, and make decision about suitable treatments and actions. Specifically, you will address the challenges of creating robust decision-making algorithms with missing data and providing clinical-based interpretability of the obtained decisions.

You will be embedded in the qurAI group, an interfaculty, multidisciplinary group between the Institute of Informatics of the University of Amsterdam and the Department of Biomedical Engineering and Physics of the Amsterdam University Medical Center (AUMC). We focus on the development, validation and clinical integration of AI solutions for data analysis challenges in healthcare. The group aims at designing and enabling socially responsible AI innovations in healthcare.

You will work under the supervision of Prof. Clarisa Sánchez from the qurAI, who will also serve as your promotor. You will also work in close collaboration with Prof. Alexander Vlaar and his group at the department of Intensive Care Medicine of AUMC. The project is defined within the Research Priory Area (RPA) AI for health decision-making at UvA. This RPA aims at conducting interdisciplinary research regarding responsible AI-driven health decision-making by joining expertise from AI, ethics, law and medicine.

You will:

  • research AI-based clinical decision-making, by developing new deep learning solutions for the analysis of waveforms obtained at ICUs;
  • publish and present your work in journals and international conferences;
  • be an active part of the qurAI and RPA group and their activities;
  • assist in educational activities and supervision of bachelor and master students.

What do we require?

  • A Master’s degree (or equivalent) in Artificial Intelligence, Computer Science, Engineering or a related subject
  • Previously demonstrated interest in artificial intelligence and medical data analysis, in the form of coursework, projects and academic publications, and an affinity with medical topics
  • Excellent programming skills, particularly in Python and ML-related libraries
  • High motivation in pursuing academic research
  • Fluency in English, both written and spoken.

Job application

If you feel the profile fits you, and you are interested in the job, we look forward to receiving your application. You can apply online via this link. We will accept applications until 11 January 2022.

Applications in .pdf should include:

  • a curriculum vitae;
  • a letter of motivation;
  • a summary of your grades;
  • any publications/theses/code you might want to share.

We will invite potential candidates for interviews within two weeks after the closing date.