The ICU Cockpit and Dashboard. How visualization and clinical decision support in critical illness can help at the bedside.

The Intensive Care Medicine Clinical Research Unit (CRU), led by Prof. Dr. Manu Malbrain, is embedded in the department of Intensive Care Medicine at the University Hospital Brussels and the Faculty of Medicine and Pharmacy of the Vrije Universiteit Brussel (Belgium). Our group combines clinical research with a large ICU (36 beds, over 2300 critically ill patients admitted each year, 10.000 ICU days, 5.000 mechanical ventilation days, and 500 renal replacement therapy days), both CRU and ICU are located on the Healthcare Campus of UZB in Jette. This unique setting allows a very fast and effective interaction among clinicians and researchers within the research team, allowing research from 'bedside to bench and from bench to bedside'. Our overall research objective is to unravel key pathways underlying critical illness-induced organ failure, in relation to compartmental pressures (intra-gastric, intra-esophageal and intra-abdominal pressure), cardio-abdominal renal syndrome, fluid and nutrition stewardship, indirect calorimetry, bio-electrical impedance analysis, less invasive to non-invasive hemodynamic monitoring, assessment of the microcirculation, thereby identifying potential therapeutic options to enhance recovery. We focus on macro- and microcirculatory underlying mechanisms of organ-specific problems evoked by critical illness. One of our research interests is the potential for helping doctors at the bedside dealing with difficult decisions using advanced computational techniques and algorithms, to allow for an early and optimal administration and adjustment of the classic ICU treatments (fluids, pressors, inotropes,…).

 

Project(s)

  1. ICU Dashboard. In modern ICU’s many physiologic parameters are validated every hour or every two hours. This can mount op easily to more than 100 physiologic parameters at each time point. At the department of Intensive Care Medicine of the university hospital Brussels, this data is electronically collected as time series of varying resolutions and integrated in a patient data management system (PDMS). The human brain can only handle or integrate a maximum of 7 ± 2 variables at a time in order to have a meaningful therapeutic decision. "The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information" is one of the most highly cited papers in psychology (1). It was published in 1956 in Psychological Review by the cognitive psychologist George A. Miller of Princeton University's Department of Psychology. It is often interpreted to argue that the number of objects an average human can hold in working memory is 7 ± 2. This is frequently referred to as Miller's law. In situations where accurate and fast decision making can be life-saving, nurses and doctors are commonly faced with a magnitude of data that cannot be overseen. In the ICU, often different types of (non)invasive catheters and tubes are placed, each connected to a different device and each monitoring different parameters and having a normal range and alarm settings. The idea for the ICU dashboard grew in 2007 as an idea for a master thesis at the Antwerp High School of project development. However, at that time it never came to a working prototype. Ever since we have been trying to identify collaborators to give birth to the ICU dashboard as the first step to support nurses and doctors with a graphical visualisation of the different parameters that were collected within the last 24 hours. Each parameter has to be labelled as low – normal range – or high with a corresponding colour going from green over orange to red. Also, the dashboard should have the possibility to adapt “normal” ranges towards the patient’s baseline conditions in order to be able to provide personalised, individualised care.
  2. ICU Cockpit. Critical Care Medicine is a relatively young branch in modern medicine. The first ICU's started in the 1950's in Denmark during the polio epidemic, but it wasn't until the 1980's that ICU's have boomed worldwide and the discipline quickly became a high-tech branch of medicine that combines clinical skills, powerful drugs, and sophisticated mechanical devices to support the function of vital organs. This allows patients to survive a variety of previously lethal insults such as multiple trauma, extensive surgery or severe infections. Intensive care is sometimes referred to as 'the art of managing extreme complexity'. Despite this dedicated care, mortality among critically ill patients who require intensive care for more than a few days remains around 20-25% worldwide. Critical illness affects millions of patients each year worldwide and consumes a large fraction of health care resources. It is therefore of great interest to detect those patients most vulnerable to specific organ deterioration as early as possible, in order to administer dedicated therapies earlier and hopefully prevent the chronic and lethal phases of critical illness. Using data mining and machine learning techniques (Random Forrest) we have previously developed clinically relevant prediction models. Together with Datastories we validated a dataset containing hemodynamic variables and treatment options. Based on simple correlations the hemodynamic variables that were used to develop the prediction model were: heart rate (HR), mean blood pressure (MBP), global end-diastolic volume index (GEDVI), extravascular lung water index (EVLWI), cardiac index (CI), stroke volume variation (SVV) and whether or not this parameter was reliable (patient must be in regular sinus rhythm and fully controlled mechanical ventilation). Furthermore, data was collected regarding underlying conditions: heart failure, septic shock, renal failure or respiratory failure or a combination. As possible treatment options 6 interventions were analysed: giving or removing fluids, increasing or decreasing inotropes and increasing or decreasing vasopressors. The aim is to determine how a large variety of measurements obtained in intensive care can be used to detect anomalous behaviour and predict relevant outcome (= therapeutic decision). At this point it still needs to be established how the measurements are related, what the criterion of interest is, how and how well different criteria can be predicted and how the resulting process can be automated. With the help of the interdisciplinary research team, where clinicians, biomedical engineers (ETRO lab) and computer scientists (AI lab) work in close collaboration, the PhD student will gradually be able to acquire the knowledge and skills to perform in-depth analysis of clinical and research databases collected at our ICU or gathered in international collaboration with other ICU's, to develop novel models for data visualization and decision support in critical illness. The ultimate goal of this project is to translate these models into bedside tools. The different research options available are:
    1. Systematic review and meta-analysis
    2. Bench-model and phantom for continuous IAP and urine output
    3. Bench-model for abdominal cavity and abdominal wall compliance
    4. Small Animal laboratory in collaboration with the Federal State University of Rio de Janeiro, Brazil (in collaboration with Prof Dr Patricia Rocco)
    5. Retrospective studies on PDMS database
    6. Prospective observational studies
    7. Prospective RCCT
  3. Monitoring parameters. Together with the visualisation the ICU department continuous the search for optimal monitoring of fluid status and hemodynamic conditions. Currently different projects in collaboration with other hospitals and industry partners are running. The PhD student will be able to collaborate in the following domains:
    1. Validation of a new continuous IBP (intra-bladder pressure) monitor. This includes a bench-top validation with a phantom designed to mimic real life situations with respiratory and cardiovascular variations and changes in compliance of the abdominal wall. It also includes a “second in human” study validation in critically ill patients.
    2. Validation of a new continuous urine output monitor. This includes a bench-top validation with a phantom designed to mimic real life situations with increased or decreased urine output. It also includes a “second in human” study validation in critically ill patients. Other points of interest are to study the relevance of the frusemide stress test in phantom simulation and critically ill.
    3. Validation of a new continuous IGP (intra-gastric pressure) monitor. Development and validation in a bench model and critically ill patients. In combination with the development of gastro-intestinal function monitoring device (gastroparesis, ileus, reflux, contractility,…). Development of plug and play device to calculate transpulmonary and transdiafragmatic pressures to be used in difficult ventilation (ARDS) and difficult to wean patients and ultimately to prevent ventilator associated pneumonia (VAP).
    4. Development of a new noninvasive IAP (intra-abdominal pressure) monitoring. In close collaboration with biomedical engineers. Development of a phantom mimicking the abdominal cavity with the possibility to change the abdominal wall compliance. Exploration of different existing techniques like tonometry (pressure gauge), bioimpedance, ultrasound, microwave reflection, photonics.
    5. Development and validation of a new double cuff endotracheal tube and controller. To prevent ventilator associated pneumonia. Second in human study in critically ill patients to prevent VAP.
    6. Development and validation of PPG sensors to monitor patients wirelessly and remotely. In order to incorporate in the Early Warning Score (EWS).
    7. Validation of bio-electrical impedance analysis. Development of new measurement technique in critically ill to assess extracellular and intracellular water as well as extra- and intravascular fluids.
    8. Development of a technique to measure total circulating blood volume in critically ill patients at the bedside. In order to answer the question whether septic patients truly hypovolemic or only vasoplegic.
    9. Implementation and follow up of fluid stewardship programmes. To provide help with the organisation of a global fluid day in analogy to nutrition day to assess for the appropriateness of the fluid prescription and administration.

Profile

The candidate should be a medical doctor (MD diploma) with a strong academic record with excellent grades, he should also already have a basic specialization diploma in internal medicine, or anaesthesia. Other qualifications:

  • Extra qualification in Critical Care Medicine is an advantage
  • Knowledge and skills in point of care ultrasound
  • Previous research experience is a plus, but not essential
  • Good track record, with at least 2 PubMed publications as first or senior author to prove the capabilities of English medical writing
  • The candidate should be willing to think out of the box and be motivated to learn and understand aspects in the field of data acquisition and monitoring.
  • Knowledge of data analysis, statistics, machine learning (knowledge of R and some programming skills) are an advantage but not essential
  • Since we are an international team, English is the main language and proficiency in written and spoken English is therefore crucial.
  • The selected candidate is expected to write a doctoral thesis on her/his research after 2 years

 

Offer

We offer a Fulltime university position for a 2-years PhD finalization in an international research team at the VU Brussels. The VU Brussels is one of Belgian’s leading research groups. The position is prefinanced for one year, and the candidate needs to apply and obtain Regional, Federal, European or other grants or fundings.

  • The Intensive Care Medicine Research Group offers a dynamic and intellectually challenging environment, in close collaboration with experts from a wide variety of other domains and disciplines.
  • A thorough scientific education, the possibility to become a good researcher.
  • The possibility to participate in international conferences and collaborations.

 

Interested?

Send your full CV with motivational letter why you are the ideal candidate amongst others. Summarize your curriculum with separate focus on education – training – skills – teaching – research – grant application. The selected candidates will be contacted to give a Skype interview where they can present themselves via a short Powerpoint presentation covering the above-mentioned aspects.

For more information please contact Mrs. Nadia Van der Steen, tel.: +32 2 477 5178, mail: secretariaat.iz@uzbrussel.be or Prof. dr. Manu Malbrain, tel.: +32 2 476 3101, mail: manu.malbrain@uzbrussel.be

You can apply for this job no later than April 30, 2020 via the online application tool:

The job starts at the beginning of the next academic year on October 1st 2020

 

Suggested reading

  1. Miller GA. The magical number seven plus or minus two: some limits on our capacity for processing information. Psychol Rev. 1956;63(2):81-97.
Wil je op de hoogte blijven van onze vacatures? Vul hieronder je emailadres in en ontvang een job-alert voor jobs die jou interesseren.
In welke categorieën bent u geïnteresseerd?
U krijgt enkel bericht over jobs in de categorieën die u selecteert