Dr Heba Khamis points the finger at the haptics conundrum
Dr Heba Khamis from the Graduate School of Biomedical Engineering has her finger on the pulse when it comes to the latest advances in haptics research.
Heba’s recent paper beat 35 other presentations to win Best Paper Award (Oral Presentation) at Eurohaptics 2014. Her work presented findings of pattern classification models applied to microneurographical data recorded from humans having mechanical stimuli applied to their fingertips.
She is fascinated by how we sense the world. Along with her research team, she is working hard to find ways to mimic nature and solve biomedical engineering problems. UNSWEngineers caught up with her to find out more.
Friction sensor prototype
Q&A with Dr Heba Khamis
What is haptics?
Haptics is everything touch-related. It’s anything from illusions and how we can trick ourselves into thinking we’re touching something when we’re not, to exploring how we actually sense and perceive touch.
What area of haptics are you involved with?
We’re interested in the fundamental physiological side of haptics and investigating what is it about the finger pads and the receptors in them that help us sense things like texture and friction. Humans have an incredibly sophisticated sense of touch and can differentiate between a huge number of objects just by touching them. We’re also particularly good at picking up and holding things with just the right amount of pressure.
We know the receptors in the finger pads signal all the information that makes this possible to the brain but we don’t yet know how, so we’re looking at the neural response to stimuli at the fingertips. We use tiny microelectrodes – the tip is something like 5 microns in diameter - and insert them directly into the median nerve that transmits the information from the receptors to the brain. The microelectrode records the neural signal from a single sensory nerve cell carrying information from touch receptors in the finger pad and then we try to decode that signal.
How did you get into this field of study?
My background is software engineering and medical science, and my PhD was in epileptic seizure detection. This doesn’t sound much like what I’m doing now, but I’m using the same principles of signal processing and pattern recognition, just applying it to different signals.
When I finished my PhD I heard that Dr Ingvars Birznieks from the School of Medical Sciences and Dr Stephen Redmond from the Graduate School of Biomedical Engineering (GSBME) were looking for a research assistant. When they described the importance of friction in our ability to pick up objects, I was intrigued and I found myself touching things and picking up random objects and wondering ‘What am I sensing? How am I doing this? I can’t believe no one’s really figured this out yet!’
What kind of real-world applications will your research lead to?
We’re looking at smarter prosthesis. There have been a few advances in recent years but they’re a bit crude and you can essentially say that all prosthetics at the moment don’t provide adequate feedback. You can control a prosthetic hand, tell it to squeeze an object, for example, but the user has no sense of how hard they’re squeezing, and whether it’s enough to safely pick up the object.
We’re trying to understand how this information would normally be transmitted so we can put feedback sensors in these prosthetics so the person can use them more naturally.
The trend at the moment is towards biomimetic design, which means closely mimicking how it happens in nature. If we can understand how it works, then we can try and replicate it, or at least it will give us a starting point so that we can design something that’s more appropriate for the particular application we’re looking into.
What’s your biggest challenge?
The biggest challenge is presenting the same stimulus with high repeatability, so we can record from one sensory nerve cell at a time but then put the signals from many nerve cells together as if they were collected at the same time.
The finger pad is a squishy little thing. If you push it or squeeze it for long enough it’s properties change completely – it might become flatter and take time to regain its plumpness. Our challenge is to apply the same stimulus even though the finger may have changed. We don’t even know it’s possible but that’s what we want to do at this stage.
Another big challenge concerns friction. No one really understands how we sense friction, because when we pick something up it doesn’t really matter how rough or smooth it is. What really matters is how slippery it is, which is not necessarily correlated with the texture.
To understand how we sense friction is likely to be the most beneficial application of all because once you know that then you can grip anything.
Are many other researchers working on this?
Not many. There’s only a few other groups in the world that can do microneurography – it’s a very niche skill set - and only one other group is looking at friction. We’re absolutely at the cutting edge of this work.
What’s next in your research?
We have just published a paper in the Journal of Neurophysiology looking specifically at decoding the signals from individual touch-sensor nerve cells, and successfully reconstructing the stimulus forces using many of these signals. This is a world first!
We’re also building prototypes of friction sensors inspired by the human finger pad. We’re looking to do more experiments with different stimuli, focusing on friction, and we’ve recently bought a new robotic arm that we can use to apply stimuli to the finger pads.