Measuring hand tremor using a pedometer
 

In my opinion being able to measure PD symptoms can help in managing your Parkinson's.

Ideally there would be a reasonably priced (<£20, say) piece of kit available that would do the job accurately. I don't know of anything. Given this, it's down to finding DIY solutions.

My present need is to measure a particularly bad left hand tremor. It's more of a flap, eminating from my wrist and causing my whole arm to shake. My fingers are moving about 2 cm with a frequency of approximately 5.5 hz.

I want to be able to measure the effect of a potential therapy. My tremor comes and goes pseudo-randomly. So, to get a reasonable comparison of before and after treatment values, I need each measurement to last many minutes and to be non-intrusive, letting me get on with life while the measurement is being taken.

My previous attempt fails to do this because you need to keep your finger on a laptop's pressure pad while the measurement is being made. See:
http://www.parkinsonsmeasurement.org/toolBox/tremor.htm

A possible way forward is to use the latest generation of pedometers. These use accelerometer technology to count steps or, in our case, shakes. They cost about £10. They can come in credit card-like sizes, which can be attached to the hand. (Alternatively, ipods, and similar products have accelerometers and apps which use these.)

The pedometer I have is almost up to the job, but:
- it has a high threshold: responds to my bad left-hand tremor, but not to the smaller right-hand tremor;
- it seems to max out at about 3 hz, so the count of the number of shakes is reduced, but consistent, one run with another;
- it doesn't give the amplitude of the tremor (but I can estimate that separately).

If you have a pedometer, I'd be grateful if you'd try it on your tremor and let us know the results.

John

some thoughts
 

John-
Regarding your attempts to devise a sensor to measure a tremor, is it reasonable to try to "get on with life" while the sensor accumulates data in the background? It seems set up to frustrate yourself. That being said, consider mounting it on the end of a staff or cane to mechanically amplify the signal.

Another possibility might be to treat your wrist as a hinge point and strap the unit to your forearm. Then extend from that to some sort of two-part sensor such as a ring coming from the relatively stable forearm and a finger tip mounted "thimble". Then a counter such as the pedometer perhaps could count the number of times that the two parts closed the circuit in a given time span. Better yet - rig some sort of recording feature and wear it as long as desired then dump it to the PC.

Finally, something that I have championed over the years and which has been totally ignored :D is the "one leg balance" where you record the length of time that you can stand on one leg without the other foot touching the floor, best out of five and each leg measured seperately. Not only cheap and portable, it can be used surreptitously as well. Gives you data on lower body strength, balance, etc.

Oh, one other thing. Are you familiar with what is called a "dance mat"? It has some real possibilities along these lines.
http://en.wikipedia.org/wiki/Dance_pad

Rick

Hi Reverett 123
I am a volunteer in a longitudinal study of the progress of PD in Oxford, UK and one of the tests was standing on one leg - timed. Another simple one which is used for Alzheimers as well was drawing a clock and the hand at I think "ten to eleven" - not quite sure. To my horror I marked the hands in wrong position and although I realised the error immediately I could not for the life of me stop being distracted by what I had done in order to correct the position of the hands and had to resort to starting from scratch drawing the clock again and then I was fine. At this point I began to have doubts as to the wisdom of volunteering

Thanks Rick and Paddy for your replies.

Please excuse the rather pedantic nature of what follows but, IMHO, testing is critical, especially if we white-rat. We need to get it right.

Timing standing on one leg to "measure" PD has a number of advantages:
- it is probably valid. I have no data or reference for this*, but I would expect the true value of one's standing on one leg time, i.e. average over many runs, correlates well with one's true PDedness, as well as with wider measures of PD such as UPDRS.
- it is understandable. We all know what it means.
- it is directly comparable between people. If you achieve 67 seconds and I 7 seconds, you're doing better than me. (Contrast this with the weight you can lift, the raw value really needs moderating by body size.)
- it is hard to cheat at. We know when it's not being done properly.
- it is easy to measure. You can use a watch.
- it can be accurately measured. With a watch you should be able to time the instant of foot down to within 1 second.
- it is, effectively, free. It uses no specialist equipment.
- it is theraputic. Doing it improves balance.
- it is relevant. It relates well to the ability to walk and dress etc..
- it is, over a wide range, linear. Doubling the time indicates double the ability.

However, at least for me, it has a number of drawbacks:
- it is open ended. On a good day I could be there for 10 minutes. When I've done this test at the hospital they've stopped it at 30 seconds: pragmatic for progression purposes, but no good if I really want to know how my condition today compares with yesterday
- it's very variable: two back to back runs can vary between 1 sec and 10 minutes. Rick I note that you use the best of 5. Certainly that reduces the variability, but it increases the time required significantly.
- there is the possibility of falling.
- it is intrusive. You can't do anything else while doing this.

And, of course, it doesn't do much to measure tremor!

[Subliminal message. Wouldn't it be useful if we agreed a set of measures and used them in white rat reports. I note I failed to do this in my recent Strobe thread.]

* Just about to post when I came across this:
"Normative Values for the Unipedal Stance Test with Eyes Open and Closed"
Springer et al.
http://web.missouri.edu/~proste/tool...inger-JGPT.pdf
Journal of Geriatric Physical Therapy, Vol 30, 1.07.
Based on the references I'd estimate it was written in 2007.

It gives averages for each age group. For instance, for people (note not just PwP) aged 50-59 based on the best of 3 trials:
with eyes open 41.2 sec;
with eyes closed 8.3 sec.

John