From ICU of the Future
Revision as of 02:24, 17 December 2010 by Larschristensen (Talk | contribs) (Draft outline for the final report)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Current group name


Group members and emails

Jeffrey Jopling – MS Health Systems and MED Student, Georgia Institute of Technology, Emory

Adriana Fuentes – MS Computer Science Student, Georgia Institute of Technology

Lars Christensen - Healthcare Technology Student, Engineering College of Aarhus

Karsten M. Jensen - Healthcare Technology Student, Engineering College of Aarhus

Current problem statement

The problem we have chosen to focus on is that hospital staff experience alarm fatigue when ICU alarms are not configured to the individual patient. Alarms are not configured at first when the patient comes in and/or they are not configured when the patients state changes during their stay.

What we experienced at the ICU after our hospital visits and conversations with staff was that some felt it was not convenient to adjust parameters, others felt it was fine, but the management gave us the impression that they were not adjusted throughout patients entire stay. We also got the impression that a suggestion from the computer on how the parameters should be adjusted would be beneficial - this suggestion should be accepted by the nurse in charge.

Description of the solution with relevant sketches

Our solution to the problem is a MatLab software where patient data will be run through to be analyzed. The data utilized for this project is from cardiovascular system, blood pressure (BP) and heart rate (HR) data.

The software developed displays BP and HR in a graph on each axe and with color mapping we will be able to add the time as a factor in the showing as the BP/HR plots changes color over time. From this data information we will be able to focus on the newest cluster of data and know where the patients steady state is and should from this information be able to make a suggestion on the thresholds for an alarm setting for each individual patient.

Focus group: Septis Patients


Plot 1


Plot 2


Plot 3


Plot 4


Work plan


Nov 22 – Nov 28

  • Working on developing code
  • Data extraction

Nov 29 – Dec 2

  • Meeting with Dr. Buchman
  • Documentation
  • Finalize code
  • Final Poster

Resources needs

  • A couple of monitors (which we can borrow here at HSI) for demonstrating the software.
  • We are using MatLab for programming, which we already have in our possession.
  • Patient data from Emory is also received.
  • Maybe a Harddrive for data storage

Main faculty coach you request

Dr. Ackerman