Epileptic Siezure Alarm

Projects General Guidance
1

I do work for a charity (REMAP) that designs and manufactures aids for people with disabilities. We're not a commercial interest, with all we build being given free to the 'client'. If a solution to a client's problem is manufactured by a commercial company specialising in disability aids we do not take on the case, but refer the client to the commercial product.

I've recently been asked if I can come up with an alarm system for a subject who occasionally takes epileptic fits.

There are detectors for use whilst a person is in bed and hence "normally at rest" but the device required is for use during a person's normal working day.

The client's fits manifest themselves by momentary unconsciousness, rather than the violent trembling that is the generally understood event. The individual then either falls to the ground or out of her chair - both of which may be considered as a period of fairly rapid motion followed by sudden stop.

Since the individual normally works, she is in constant motion so a simple motion detector cannot be used as a sensor.

Tossing ideas around I thought about a motion sensor (accelerometer) attached to a processor that would interpret a movement (vertical) followed by a sudden stop as a "fit" and sound an alarm.

Has anyone out there already developed such a device for either this or some other application that might be adapted to suit this topic.

Alternative ideas and/or concepts would be most welcome.

Many thanks

2

Does her heart rate change?

3

Many thanks for the fast response.
Since this is a "passed on" enquiry, I have no knowledge of the client at this time.
However, a Google search does suggest abnormal heart rythme might be a means of detection so will pursue this line of enquiry.
regards

4

I can think of two potential problems with using motion...

• What if the client does not fall out of her chair? Presumably that sometimes happens.

• Is it possible to tell the difference between a fit and stopping to pick up a dropped object?

5

The first problem you identified is a real one.

The second one I anticipate might be resolved by the difference between a controlled voluntary movement and a sudden stop caused by hitting the floor. I surmise it is physically impossible for the human body in motion to effect immediate stoppage (physical dynamics), whereas hitting the floor (an immovable object) would cause sudden stoppage.

But your suggestion on heart rate will be investigated as the first possibility, since it could cover the eventuality of the "not-falling-out-of-the-chair" scenario; it's much more elegant and probably requires less power to provide a functional monitor.

6

I suspect your best course of action would be to consult a doctor to discover what are the safest and most reliable detection methods. Preferably, the doctor who is managing this patient's epilepsy.

7

OP, it's not quite clear. Do you want something that signals other people when the event
occurs, or attempts to signal the person undergoing the event [if that would help preclude
it]?

Also, how long do they stay out?

This sounds a lot like narcolepsy, from what I understand it to be.

8

jackrae:
There are detectors for use whilst a person is in bed and hence "normally at rest" but the device required is for use during a person's normal working day.

Do you know how these detectors work, i.e. what do they sense and key on? Or can you give a link to the maker of these detectors? I know they don't cover the scenario you're interested in, but still may be a starting point.

9

What about dogs? There are different kinds of epileptic siezures.

10

Taking a step back to describe the scenario.

Firstly, all of the comments are good and certainly provide both food for thought and options; many thanks.

The original enquiry came from a friend who advised the 'client' worked in an office and occasionally experienced seizures, exacerbated by the work-place environment (office heat). So the firm gave her a nice cool office of her own which eased the occurrences but now her work-place colleagues are not readily aware if and when she takes a seizure.

My immediate response was that the company still has a duty of care towards all its employees and if the seizure is liable to cause an injury, or worse, then she should not be permitted to work unaccompanied. My friend is giving this some consideration !

For a previous case-client I built a pram-brake system which would automatically apply the brakes if she experienced a seizure, fell and released the pram. Having been asked by my friend on ideas for his work-place colleague I am interested in a looking into a general every-day sensor system. If the body can give a sub-conscious pre-warning of a seizure (heart-rate ?) then the individual may have time to take action to either eliminate of minimise the effect of a seizure.

So the original enquiry role has changed from that of a specific case to a general one on how to build a detection system

11

Seems rather difficult...

Best way for electronics to detect a seizure would be wearing an EEG-monitor all day.
Having E. myself, I guess that probably isn't an option, it isn't comfortable to wear a cap
all day and you might get false readings every now and then. One ofcourse also needs to be
familliar with brainwave patterns to be able to read a pattern as seizure.

Never the less, in case you're interested, it is possible to build an diy EEG-monitor
http://openeeg.sourceforge.net/doc/

I don't know whether it is a good idea, perhaps you could do something with RFID. By more or less
paving the floor with rfid tags and giving her a sensor in the form of a necklace, you could detect
that she's close to /laying on the ground. That would ofcourse not work when she gets a seizure somewhere else and expecting someone to stay put in the same room from 9-5 or limiting her acces to specific RFID-enabled locations might change her workingplace into a "prison".

It might also be possible to let her play an active role, but I'm not fond of that one as well. When she has a seizure her mind is absent and by letting her push a button every 3 minutes a system could detect she doesn't have a seizure. But once again, it's inviting false readings, when she doesn't press the button for some other reason, her co-workers may think she has a seizure.

One could prevent that by letting the device beep after 150 seconds warning she still has 30 secs
to push the button. I don't know how frequently she has seizures, pushing a button isn't hard, but if you have to do it every 3 minutes and have a seizure once a month, she's confronted with her epilepsy day in day out.

As suggested I would not just make a system, but contact her neurologist as well, look at legal
implications and perhaps get in contact with an E-support group as well.

Getting an official "medical approved" label for a device you want to make probably costs a factor 1000
more as the costs of components, design, programming and assembly. First goal is of course to help her, but I... would like to be a 100% certain a device like that won't result in a giant lawsuit from one of all parties involved.

Finding someone who also feels comfortable in a colder environment and would work side by side would imo be the best idea.

One thing that wonders me is how she... thinks of working alone and what the real danger of working
alone is in her case. I am fortunate to have all my seizures in bed. Some people with E indeed can't
work alone because a seizure might even kill them, but I also know others that have less severe seizures, avoid stuff like heavy machinery and think the occasional bump on the head is a very small price to live a normal life.

12

Does her office have windows where co-workers can watch her?

13

working in an office often means using a computer
would it be possible to monitor the computers state (active or idle) along with the above mentioned tilt and impact sensors?
beyond that as i have found with my own battle and some research the opical lobe of the brain is often a good point to moniter if you can get proper (safe) electrodes and wire them correctly

14

What do you mean by computers state Idle?
If a person is in a meeting for an hour or so, th computer might be idle but no alarm should go off imho.

15

in a meeting there would be others to monitor ... no need for additional help..
such a thing can be turned off when around others who will see and recognize arising problems

again this is only what has seemed to work for me and my computer in battles with E.

16

I know, this thread is old and the problem has probably already been solved in one way or another, but I just wanted to remark that there are devices called "dead man switches" on the market for ages. We use them in the lab I'm working at in case someone needs to work alone or out of sight of his colleagues for a longer time.
They are little devices which detect the position of the wearer. You clip them to your belt and if you fall over and lay on the ground for a certain time, they sound a so-called pre-alarm which you need to acknowledge. If you don't, the main alarm is forwarded to the security guys and they come looking for you.

17

I arrived at this thread while searching for the solution of a similar issue. I guess you can give this a read.

http://faer.ac.in/motorola/msp2007_08/05_Smart%20Epilepsy%20prediction%20system.pdf