EPFL/STI/IEL/LTS2
Expert in Image processing
Pierre Vandergheynst
EPFL
NewbornCare
Reducing the false alarms of neonate vital sign monitoring via a computer vision-based approach to accurately measure heart and respiratory rates in a contactless way
Printable Summary in PDF


CSEM
Expert in biomedical signal processing and in particular multiresolution tools for use in bioimaging applications
Amina Chebira
CSEM
USZ/Neonatology - University Hospital of Zurich
Expert in neuroprotection in extremely premature infants, cerebral circulation during neonatal adaptation
Jean-Claude Fauchère
USZ
CSEM/Signal
Expert in dynamics of cardiac systems using stochastic pacing and frequency domain analyses
Mathieu Lemay
CSEM
EPFL/STI/IEL/STI-JMV
Expert in adaptive filtering, adaptive frequency tracking, robust spectral estimation, nonlinear signal modeling, electrocardiogram
Jean-Marc Vesin
EPFL
USZ/Neonatology/BORL, ETHZ/NCZ
Expert in miniaturised systems for the life sciences, clinical trial in NIRS Spectrophotometry
Martin Wolf
USZ

Project Description

Modern societies are giving more and more priority to increase the quality of neonatal and post neonatal health sectors. The actual neonate sensor setup for heart and respiratory activities, and oxygen saturation is the following: gel electrodes to monitor the heart rate by electrocardiogram (ECG) and pulse oximeter to monitor arterial oxygen saturation (SpO2). The combination of the increasing number of parameters being monitored and the sensitivity of these sensors to body movement (especially the limbs) is responsible for the inacceptable high rate of false alarms, which in turn generates discomfort, stress and cardio-respiratory instability. These false alarms may also be the cause for caregiver desensitization which may dangerously lead to long response times for true alarms. Even if the problem is well known, it has not received new major incomes from the scientific community. 

[read on]
Read the Project Presentation

Our researchers in the media

Notable publications

2015 3DTV-Conference: The True Vision-Capture, Transmission and Display of 3D Video
Y. Schoenenberger, J. Paratte, P. Vandergheynst
2015 3DTV-Conference

Graph-based denoising for time-varying point clouds
Y. Schoenenberger, J. Paratte, P. Vandergheynst
3DTV-Conference: The True Vision-Capture, Transmission and Display of 3D Video

A multimodal approach to reduce false arrhythmia alarms in the intensive care unit
Sibylle Fallet, Sasan Yazdani, Jean-Marc Vesin
Computing in Cardiology 2015, Nice, France

Adaptive frequency tracking for robust heart rate estimation using wrist-type photoplethysmographic signals during physical exercise
Sibylle Fallet, Jean-Marc Vesin
Computing in Cardiology 2015, Nice, France

Posters from 2016


Camera-Based Respiration Monitoring
Fabian Braun, Alia Lemkaddem, Virginie Moser, Stephan Dasen, Olivier Grossenbacher

Pipeline for real-time heart-rate estimation from video streams
Sibylle Fallet, Lionel Martin, Yann Schoenenberger, Jean-Marc Vesin, Pierre Vandergheynst

Posters from 2015


Reduction of False Alarms in Neonatal Vital Sign Monitoring
Daniel Ostojic, Stefan Kleiser, Nassim Nasseri, Martin Wolf

Monitoring of heart and respiratory rates in newborn infants using video sequences
Sibylle Fallet, Yann Schoenenberger, Jean-Marc Vesin, Pierre Vandergheynst

Posters from 2014


Newborn Monitoring based on Multiple Vision Sensors
Sibylle Fallet, Jean-Marc Vesin, Pierre Vandergheynst, Martin Wolf, Jean-Claude Fauchère, Mathieu Lemay, Amina Chebira, Damien Ferrario

    

Project Photos