TexasTom thanks for sending the screen shots from the "Emma".

I ran my prototype (that is with only 4 vibrators) using the firing pattern shown by the screen shot. No luck. I show the code below.

I reckon that the "Activator" that they talk about is just the coin shaped vibrating motor that we are using.

Activator 1 is shown as having a firing pattern alternating between on (100ms) and off (100ms) giving a period of 200ms, or a frequency of 5Hz. That's in the middle of the range of 4-6Hz that's the norm for PD. I don't think this is a coincidence.

Given that tremor varies from person to person, it makes sense to try a range of values for the period from about 150Hz to about 250Hz.

With this perspective it is possible that activators 2 to 6 are just dealing with errors. These parameters are possibly determined in a training session. Here's how I could see such a system working:

- use a sensor to measure tremor;
- accept the alternating pattern of Activator 1, but play with the period until the locally optimal period is found for this particular user;
- take measurements with Activator 1 set, where there is still tremor set the Activator 2 bit to on, possibly with a reduced strength;
- and so on, giving less strength to each succeeding activator.
(Basically a Fourier analysis type approach.)

The moral of the story is that unless the tremor reduction system can be got working by trial and error, a way of measuring tremor (accelerometer, etc.) is at the heart of the system.

Ironically, I wanted to use this time for further tests, but for some reason although I'm late with my drugs I'm not tremoring!


// NOTE: this is a prototype. DO NOT USE FOR CLINICAL PURPOSES.

// This program controls 6 vibrating motors positioned in a circle around a wrist or arm.

// The theory under test is that the vibrations around the wrist reduces tremors.

// D0 A5
// D1 A4
// D2 A3
// V0 <---- D3 A2
// D4 A1
// V1 <---- D5 A0
// V2 <---- D6 ARDUINO
// D7 UNO VIN
// D8 GND
// V3 <---- D9 GND
// V4 <---- D10 5V
// V5 <---- D11 3.3V
// D12
// D13
// USB 5V
//

const int NVIBRATORS = 6;
const int PHYSICAL_PIN[6] = {3,5,6,9,10,11}; // Choose digital pins with PWM.
const int INTERVAL = 100; // ms
const int NINTERVALS = 20;
const int STRENGTH = 200; // range 0..255
const int VIBRATOR[6][20] = {
{1,0,1,0,1,0,1,0,1,0,1,0,1,0,1,0,1,0,1,0},
{0,1,0,0,1,0,0,0,1,0,0,0,1,0,0,0,1,1,0,0},
{0,0,0,1,1,0,1,0,1,1,0,1,1,1,0,1,1,0,0,0},
{1,0,1,1,0,1,0,0,1,0,1,0,1,0,0,0,1,0,0,0},
{1,1,0,1,0,0,1,0,1,0,0,1,0,0,1,1,0,1,0,0},
{1,0,0,1,0,1,0,1,0,1,0,1,0,1,0,0,1,0,0,1}
}; // 1=on, 2=off

int intervalNumber = 0;

void setup()
{
// set the vibrator pins as output
for (int v=0; v<NVIBRATORS; v++)
{
pinMode(PHYSICAL_PIN[v], OUTPUT);
}
}

void loop()
{
for (int v=0; v<NVIBRATORS; v++)
{
analogWrite(PHYSICAL_PIN[v], STRENGTH*VIBRATOR[v][intervalNumber]);
}
delay(100);
intervalNumber = (intervalNumber+1)%NINTERVALS;
}


John

I suspect you meant a period of 150mS to 250mS

I'm in the midst of trying to get a "cubby hole" built for our boxing class. Everyone tosses bags on the floors, old style gym lockers are too small, so I'm doing a little weekend wood working project. Taking longer than I realized as I hadn't been in the workshop for a couple of years. Great fun to actually have a fun project, but will get these knocked out of the way then back to the Emma.

Tom

Its great to see the ball rolling in attempts to develop a device to reduce tremor. My wrist tremor is abated if my hand, wrist or arm scrapes against something so this concept of using sensory nerves to reduce tremor fits with my experience.

The Arduino could have a lot of potential. How is this to detect the tremors? Can the micro vibrating motors be used as motor-generators and thereby detect the motion of the tremor? Or are there plans to incorporate an accelerometer? Arduinos run a limited version of C++. Is that expected to be the language of choice? I suppose there are many ways to put together an algorithm that would respond to the 4-6 Hz motion of tremor but not to other things.

Bill Patterson

Bill,

My original thought was just to try the motors. The screen shot shows when each of the six motors is active. Just to get an idea if it works.

Currently no tremor measurement is done, but thankfully I also have a MM7150 Sensor Fusion Module that can measure all sort of things.

Post DBS I had zero tremor. I still have zero tremor in my left hand, but right at times have difficulty with a keyboard and using a mouse. Sigh. Still working on dialing in my DBS.

I built a copy of the Emma Watch and tried it out on myself. Despite trying many frequencies and patterns, it had no effect whatever on the Parkinson's Disease tremor in my left hand. Despite a diligent search, I haven't found any evidence of the Emma Watch working on anyone besides Emma.

Microsoft came out with the Emma Watch with great fanfare in 2017 and they've obtained a patent on it but no further results have been released. The media seem to have uncritically accepted the premise without proof that the Emma Watch will help PD patients with tremor. I'm concerned that Microsoft has raised a lot of potentially false hopes in many people with PD.

I posted a YouTube video "Does the Emma Watch work for anyone besides Emma" about my experiments with a copy of the Emma Watch that I built. It's not hard at all to make and I plan on posting another video on how to make one.

cbluesdog,

Welcome to the forum.

The link to the video that you mentioned is:

Does the Emma Watch work for anyone besides Emma? - YouTube

John

Microsoft still, in 2019, considers this project in the research stage only (Project Emma - Microsoft Research). I'm wondering why no one has yet (or maybe I missed it) publicized an app for a watch to vibrate when its sensors detect a tremor. Does anyone know more about this?

Where do I buy one? NEVERMIND......