Program 2019
MAY 15, 2019. ANTWERP. BELGIUM
Make Artificial Intelligence work
for your business
11:00
12:30
Registration & demo exhibition
12:30
14:00
Lunch & demo exhibition
14:00
14:15
Sami Mahroum - Director Research & Strategy, Dubai Future Foundation
The AI Discussion We Need
14:15
14:30
Jan Rabaey - CTO System-Technology Co-Optimisation, imec
The Cognitive Edge
14:30
14:50
Jo De Boeck - Chief Strategy Officer, imec
Addressing grand challenges in AI in Flanders’ Top-Research program
15:00
17:30
15:00
15:05
Introduction
15:05
16:10
Program: Part 1
Hans Constandt - CEO, Ontoforce
Hans Danneels - Co-Founder & CEO, Byteflies
Roel Wuyts - Principal Scientist, imec
Liesbet Lagae - Program Director Life Science Technologies, imec
Life science chip technology meets artificial intelligence
Pieter Peeters - Senior Director Computational Biology – Discovery Sciences, Janssen Research & Development
From Patients to Insights, to novel breakthrough Therapies
Q&A
16:10
17:30
Program: Part 2
Anton Vedder - Professor of Law & IT, KU Leuven
Towards ethically responsible AI in a medical context
Isabelle Dehaene - Adjunct Head, Women Hospital UZ Gent & Sofie Van Hoecke - Associate Professor Semantic Intelligence, IDLab, imec Research Group at Ghent University
Machine learning on data from medical records and wearables to predict preterm birth risk
Michiel Rooijakkers - Lead Signal Processing, Bloomlife
Machine learning on data from medical records and wearables to predict preterm birth risk
Tom Oostrom - Managing Director, Dutch Kidney Foundation
How technology can empower kidney patients to lead their life
Chris Van Hoof - Vice President Connected Health Solutions
How technology can empower kidney patients to lead their life
Georges De Feu - CEO, LynxCare
ROI of Big Data in hospitals
Q&A / Closing remarks
17:00
19:00
Closing reception
location: demo floor
talks by representatives
from local industry and organizations.
live demos
of ground-breaking technologies & applications.
Receive relevant updates via email
Speaker:
Presentation abstract: Machine learning on data from medical records and wearables to predict preterm birth risk
Each year, 1 million babies die as a result of preterm birth complications. Yet, many of them could be saved — provided timely and well-chosen medical interventions. What if we could help caregivers in their treatments, by predicting whether and when patients will deliver preterm?
Preterm birth is, with a worldwide incidence of 5 to 18%, a major contributor to neonatal morbidity and mortality. Preterm birth can be the result of spontaneous preterm labour, preterm prelabour rupture of membranes or other obstetrical pathology that warrants delivery. Timely management is crucial.
In order to prevent under- and overtreatment, it is important to differentiate patients who will deliver prematurely and those who will not. Several data-driven prediction models have been proposed to tackle this problem. Nearly all of these studies apply logistic regression models. The use of more advanced machine learning techniques (e.g. gradient boosting, deep artificial neural networks) has not been explored in this area, nor has the use of interpretable machine learning models been proposed.
In the first part of the presentation, we show how AI can shape the future of health, especially to improve timely medical interventions. By combining clinical observations and freely written doctors’ notes, the unique machine learning models of IDLab and Ghent University Hospital predict the chance of giving birth within the next seven days. This way, the therapy can be adjusted accordingly to optimize outcome for mother and child.
In the second part of the presentation, we show how physiological data collected with a wearable sensor can be used to detect early onset of preterm labor, which could potentially help refine the prediction.
Biography
Michiel Rooijakkers earned his Master degree in Electrical Engineering from the University of Technology Eindhoven (TU/e), in 2010 for his work on ultra-low-power R-peak detection at Philips Research. He continued his research at the Signal Processing group of the TU/e, where he obtained his PhD for his work on “Signal Analysis for Continuous Electrophysiological Monitoring of Pregnancy”. The focus in this multi-disciplinary research topic involves applying, implementing, and optimizing signal processing techniques for low-power wearable systems to enable continuous ambulatory monitoring of the maternal and fetus health and the progression of pregnancy using sensors placed on the maternal abdomen. Since 2014 he has been part of the Bloomlife team in the position of Lead Signal Processing, where leads the team`s efforts to unlock new clinically relevant features and functionality to improve people’s quality of life around pregnancy and birth using smart at-home ambulatory monitoring techniques.