Frans van der Helm

‘We use force perturbations generated by a robotic arm to measure the neurological effects of a stroke.’



Frans van der Helm earned his PhD in 1991 from TU Delft. He has been Professor of Biomechatronics and Biorobotics at TU Delft, Faculty Mechanical, Maritime and Materials Engineering since 1999 and he also has appointments at LUMC, the University of Twente, Case Western Reserve University (Cleveland) and Northwestern University (Chicago). He is programme leader of two STW research programmes: NeuroSIPE (system identification and parameter estimation of neurophysiological systems) and H-Haptics (Human Centered Haptics). In 2011, he was awarded an ERC advanced grant in the field of 4D EEG, for research on the spatial and temporal activity patterns in the brain. In 2012, Frans van der Helm was awarded the STW Simon Stevin Meester prize for outstanding research in technical sciences.


TU Delft: Biomechanical Engineering


Theme leader Interventions and Care
Technical Scientific Leader Neurocontrol (IMDI)
Coordinator BSc programme Clinical Technology

Using a robot to measure neurological effects

Rehabilitation advice after a stroke

‘Areas of the brain in stroke patients, particularly the motor or sensory cortex, are often damaged. We want to under- stand how the brain adapts to such damage – and use this information to provide advice for appropriate rehabilitation.’

‘After a cerebral hemorrhage other parts of the brain take over functions. Functions that were damaged on the right side of the brain are often taken over by the corresponding area on the left side of the brain. So, for example: if the right brain hemisphere that controls the left leg is damaged, a part of the left brain hemisphere will take over the control of both legs. This will result in limitations because there are no direct pathways to the muscles, for instance, which means that modification of reflex- es works less well. When we exe- cute a movement we can modify our hand or foot ‘in a reflex’, but this will be difficult for people where the shift and ‘stacking’ of functions has taken place.’

Using a robot arm to measure neurological effects

‘In order to measure the neuro- logical effects of a stroke, we use force perturbations generated by a robotic arm. The patient must assume a certain position or carry out a strength task and while doing so various measure- ments will be made, such as the position of body parts and the stiffness and electrical activity of particular muscles.’

‘Furthermore, we also measure the electrical activity in the cortex using an EEG. By collating the measurements the reflex strength can be calculated for the position, velocity and force feedback. This allows us to enter a wide variety of conditions and observe the effect inside the central nervous system. This supports the hypothesis that the main activity, the regulation of reflexes, occurs in the cortex.’

Rehabilitation therapy based on the changes in the brain

‘We want to understand what happens in the brain during the first weeks after a stroke: does the damaged part recover and does it resume its original function? Or has this function permanently ‘moved’ to the other brain hemisphere? This is very important as each case needs a different rehabilitation treatment: if the original part recovers and resumes its functions, the patient must re-learn movements through repetition. This works best. If the other brain hemi- sphere takes over this function permanently, a lot of variety should be provided so that the part of the brain that is now dou- bly burdened learns to cope with the extra task.’

‘The term ‘cerebral hemorrhage’ is actually incorrect. In 90% of the cases it is an infarction, in other words, it means that there is insufficient blood flow to the brain and poor oxygen supply, which subsequently leads to the death of brain cells. This damage cannot be seen on an MRI scan, or it is seen too late. What we want to be able to do is establish at a very early stage which areas of the brain have been damaged by the infarction, and whether or not the damaged parts will recover. We want to know this as early as possible so that appro- priate therapy can then also be started as quickly as possible.’

Collaboration in the Medical Delta region

‘The technical knowledge, the ro- botics, signal processing ... this is where Delft’s expertise lies. We work closely with the Neurology and Rehabilitation departments at LUMC and the department of Rehabilitation at Erasmus MC. In addition, we work in close cooperation with VUMC because of their expertise in the field of EEG.’

‘We have already conducted a few brief studies in which we concentrated on data acquisition and modeling; after the summer we will start with the first large trial with patients. To be able to recruit enough patients we will collaborate not only with the medical centers within the Medical Delta alliance, but also with many other non-academic hospitals in the Netherlands, from Medisch Spectrum Twente in Enschede to Reinier de Graaf Hospital in Delft.’