[johnt]

If we want to understand Parkinson's or to be able to judge the effectiveness of potential therapies, we must measure symptoms. The issue that I want to address here is the frequency with which we take measurements.

Little et al. report [1]:
"Currently, there is a scarcity of objective clinical information about symptom dynamics at intervals shorter than 3 months stretching over several years, but Internet-based patient self-report platforms may change this."

While the measurement of symptoms, e.g. UPDRS, requires costly clinical involvement and time consuming attendance at a medical centre there is understandable reluctance to repeat the test frequently.

With automated, unobtrusive measurements, and with data communicated over the web, it is possible to take measurements every second or even more frequently.

To a certain extent getting more information off of a few people is equivalent to getting less information off of more people.

For instance, knowing that the performance of one person improved by 12% over a year, is less powerful than knowing that performance improved by 1% in each month of the year.

Reference

[1] J Med Internet Res. 2013 January; 15(1): e20.
"Quantifying Short-Term Dynamics of Parkinson’s Disease Using Self-Reported Symptom Data From an Internet Social Network"
Max Little, Paul Wicks, Timothy Vaughan and Alex Pentland

John

[Tupelo3]

John, you may be interested in a phone app that you can download that measures hand tremors. Its called LiftPulse. I have been using it during the day to follow my tremors. Sometimes shows nothing when I'm relaxed with a glass of wine or book. Other times goes very high like when I'm trading the stock market or watching an exciting sporting event. My tremor is clearly related to both stressful and adrenaline rush situations.

[johnt]

Tupelo3,

The app you refer to sounds interesting. Does it provide frquency information, can it run unobtrusively 24/7, can you download the results for further analysis? I understand that you have a background in stats given this could you run a spectral analysis of your data. This will probably show a peak at about 5hz but it may also contain peaks at lower frequencies.

I've written an on-line tremor measurement tool. It works by tracking the movement of the mouse pointer that arises if you put a finger on a laptop's pressure pad or put your hand on the mouse. Unfortunately, it only works with moderate degrees of tremor and doesn't lend itself to 24/7 service.

See:

http://www.parkinsonsmeasurement.org/toolBox/tremor.htm

John

[Tupelo3]

Originally Posted by johnt

Tupelo3,

The app you refer to sounds interesting. Does it provide frquency information, can it run unobtrusively 24/7, can you download the results for further analysis?

John

John,

Unfortunately, the app cannot run by itself. You have to tap a button and hold the phone for 10 seconds. The app is calibrated to measure tremor frequency and amplitude. The tremor peaks tend to be between 5 and 10hz. The app sensors tremors in all directions (x,y,z) and calculates an overall magnitude amplitude in centimeters.

Its basic function is to first set a baseline and then it measures future readings against the baseline score. I have no idea how accurate the measures are, but it certainly works well for following trends versus the baseline.

For anyone interested, the app is free at this time
Lift Pulse

[soccertese]

will open up do it yourself clinical trials and with finger print readers, researchers can verify identity of participants.

heck, you can take a picture of your taking the supplement, experimental drug being tested, might allow more participants who can't take off work or travel far distances to participate in trials.

http://medcitynews.com/2013/09/iphon...ain-for-apple/

[Debi Brooks]

MJFF is developing tools to collect data (with varying frequency) to be pooled and shared with researchers. We see the greatest value in patient profiles that link genotypic information with phenotypic information (especially objective data that reflects disease variability, on and off meds overtime). We believe this data will be of great value to patients and their physicians for disease management and we know it will be transform our understanding of risk factors and disease course of PD. Which is ultimately needed to develop better treatments for all aspects of PD.

We are working on tools where data would complement detailed profiles (dna, patient histories, medications, symptoms etc) to provide a complete picture and we are counting on the fact that patients will be willing play the pivotal role in providing these data.

We also know that data privacy, data integrity, data access, and data sharing requirements for such a project are non-trivial. Stay tuned as this project rolls out in 2014 as an adjunct to Fox Trial Finder.

Debi

PS: If you haven't created a patient profile in our online clinical trial matching tool, Fox Trial Finder, please check it out. www.foxtrialfinder.org

[johnt]

Debi,

That's good to hear.

One of the key factors for anyone active in this area to consider is the platform used to collect the data. Going from more patient coverage to less but, conversely, from less measurement power to more, we have:

1. on-line tests, run directly off the internet, requiring no installation. This is the route I've gone down. It let's you do tap tests well, but tremor tests less well.

2. native apps running on a PC gives you more powerful tests because it gets you closer to the hardware. I chose not to do this because I expected that people would be reluctant to install programs written by an unknown person. I think this is less of a worry for an organization such as MJFF with good "brand" trust.

3. smart phones and tablets, these allow a wider range of tests because they come with onboard sensors such as accelerometers and GPS. They are also "wearable" so can collect data almost 24/7. Information such as distance and speed of walking can be collected.

4. bespoke hardware that measures a wider range of signs, such as rates of swallowing, blinking and drooling.

An important distinction between the methods is one of cost. Options 1 and 2 are effectively free for a large number of PwP, because they already own a PC or laptop. I suspect that smart phone ownership by PwP is large and is growing. Option 4 has a cost which it is likely will be have to be paid for by PwP themselves (unless someone is prepared to pay for it in a similar way to the generous payment by Sergey Brin for 23andMe genetic testing for PwP).

How much would people be prepared to pay for bespoke hardware?

Finally, it seems to me, that the really exciting step is the one that follows this: high frequency measurement of thousands of people, with integrated reporting, allows high frequency clinical trials, giving answers to important questions in weeks, not years.

John

[Debi Brooks]

There are many elements at play as we have jumped in to the deep end of this pool on this. Some highlights of our thinking...

• We can currently collect (cost is the only limiting factor) sufficient biological data (terabytes needed) to be useful (DNA, biomarker profiles in blood and CSF)
• We are watching as costs for sequencing come down so we can activate on the largest population possible
  
• Unfortunately we can't correlate big data on the biology side with the limited, small data available on the clinical side (clinical scales such as UPDRS aren't nearly sensitive enough)
• We are evaluating the over 200 tracking devices commercially available to understand which will be most relevant for measuring PD patient activities. Such devices could collect big clinical data objectively (which is our goal) but need to validated as predictive tools
  
• Our recently formed data science team is working with elite technology partners to harness these possibilities across all of our significant data sets.
  

Debi

[johnt]

Thanks for this Debi!

As someone 8 years after diagnosis, time is important. So I hope that this excellent MJFF initiative goes live as quickly as possible.

Every member of this forum could, in my opinion, help expedite the work of MJFF, and others working in this area, by:

1. Suggesting objective measurements that most closely match their overall experience of PD. Ideally these should not require a conscious intervention by the PwP, e.g. stoop or walking speed.

2. Enter data on PDMeasure:

http://www.parkinsonsmeasurement.org/PDMeasure

This is a web site that I run which allows people to enter data, such as month of diagnosis, and to take an online test, which is similar to one in the CAPSIT-PD set of tests.

There are currently 102 people enrolled.

If you haven't already enrolled, please register now. It takes just minutes to enter data and take a side-to-side tap test. The data is stored anonymously, but is open to anyone to analyse.

If you've already enrolled, please take the tap test frequently. Long sequences of measurements spanning months, or even better years, are especially valuable, because they help us to validate or, on the contrary, to invalidate the test.

I'll pass any knowledge that we gain to MJFF: it might just help them in a small way.

A final point, high frequency measurement enables a more personalized form of medicine. It may be, for instance, that a trial of a new therapy does not show any statistically significant change, for better or for worse, across the whole cohort. But, who would argue with a person continuing, well monitored, of course, with the treatment who across 500 daily measurements before the trial had shown a steady decline, but in the 200 days, say, of the trial had shown a steady improvement?

John

[vlhperry]

*** I have mentioned time and time again, I have 2 mutations in my Parkin gene; deletions of exons 3 & 5. My doctor has been seeing me quite frequently as of late in an attempt to see if my DBS surgery has been beneficial or a hindrance to me.. My batteries are over 8 years old and show signs of weakening. He has determined my left side is being helped but is concerned that my right side is being helped. I am to go in this Monday, 2 weeks from the last time I was in
and am to be off meds. He turned off my right side already. I know they need the information, but I am tired of being a guinea pig.

Dianna