Louis Atallah

I will be moving to Philips Netherlands in Feb 2012 to take up a post as a senior scientist in patient care solutions and won't be maintaining this page. For any info, please use my Philips email.

I am currently a Research Fellow in the Pervasive Computing group, part of the
Hamlyn Centre at Imperial college, London working on applications of machine learning in Body sensor networks and surgical skill evaluation.

I work on the use of sensing for behaviour profiling, observing how people learn and how their behaviour changes over time. Our team at Imperial is currently working on a combination of ambient and wearable sensors that can help track users, classify activities and observe transitions. We are also investigating the use of an ear worn sensor for long term activity profiling, used in COPD, orthopaedic and elderly care studies.

Previously, I finished my Dphil at the robotics group in Oxford, under the supervision of Dr Penny Probert Smith. The main aim of my Dphil was the application of Machine learning techniques for underwater seabed classification.  I then worked as a Lecturer in the British University in Dubai in collaboration with the University of Edinburgh

Recent Projects

ESPRIT: Elite Sports Performance Research in Training

New paper: Pervasive sensing for athletic training B.P.L. Lo, L. Atallah, B. Crewther, A.M. Spehar-Deleze, S. Anastasova, P. Conway, C. Cook, S. Drawer, P. Vadgama, and G-Z Yang, in Delivering London 2012: ICT Enabling the Games, The IET special interest publication, Nov 2011, pages 53-62.pdf

e-AR: An ear worn sensor for healthcare and sports monitoring


Winner of the Bluetooth World Cup 2010 (among 270 international teams) also winner of the healthcare category.
Finalist in the IET Awards 2010.
Winnner of the medical futures award for research translation in ENT

Inspired by the function of the human inner ear, the ear worn sensor is a small device that fits discreetly behind the ear and captures information on the  balance of the wearer. With advanced signal processing, further information about the gait, posture, skeletal/joint shock-wave transmission and activity of the individual can be deduced. The e-AR sensor could also detect sway, unstable movement and the risk of falling. Data is transmitted wirelessly to a receiver that can be connected to a PC, a PDA or a laptop. Related media articles include:
New Scientist, BBC, Video: Available here.

Related Papers:

Energy expenditure prediction using a miniaturised ear worn sensor -Medicine & Science in Sports & Exercise: July 2011 - Volume 43 - Issue 7 - pp 1369-1377

Observing Recovery from Knee-Replacement Surgery by using Wearable Sensors, L. Atallah, G.J. Jones, R. Ali, J. Leong, B. Lo and G-Z. Yang. in BSN 2011 in Dallas, Texas.

Real time pervasive monitoring for post operative care, BSN 2007


latallah (at) doc.ic.ac.uk

Department of Computing, Imperial College, London