job1119,

Point taken, but why let this stop us?

The problems are many:
- perhaps the gadget only works for Emma's version of PD;
- perhaps the response seen was only a placebo (but, let us not write off placebos, they have a value);
- perhaps the gadget has more complexity than we think (before this programme was broadcast I thought the solution was to fire tiny electrical shocks at the arm so as to cancel out the signals causing the tremor).

Progress has been made in similar areas: the non-spill spoon; a gyroscope strapped to the wrist.

I like the ethos of the show: you have a problem, we can likely reduce it. Or, taking it one stage further, I have a problem, I can likely reduce it.

The rapid development approach adopted by this programme, with its "can do" attitude, contrasts with the Big Pharma/ Big Medicine/ FDA approach of creating hurdle after hurdle to climb over.

John

As far as I know no further details of the "Emma" have been published. So, I've gone ahead with my own ideas of how vibrating motors may reduce tremor.

I've built my first prototype using 3 flat vibrators.

The device runs off an Arduino Uno, which is connected to my laptop. This allows me to have better control of the settings, but at the cost of being tethered to the machine. It will be easy to make the device stand-alone, with the electronics about the size of a pack of cards, and later down to watch size. The vibrators (diameter about 8mm and height about 4mm) are positioned on a velcro strap around my wrist.

The user experience is pleasant, and you quickly forget that it's there. The device makes a noise. I find this OK, but other people might object.

Does it work? I don't know. I've convinced myself that with a bad setting (e.g. 5Hz) it makes my tremor worse. I've not managed to find any setting which has definitely causally reduced my tremor, but 0.5Hz and 10Hz look promising. Even with the placebo effect working to my advantage, I've been unable to find any setting which has stopped all tremors. However, and it may be just coincidence, I've had many cases where I've changed the frequency and the tremor has reduced within seconds. Unfortunately, because the tremors are so variable this does not prove that the device works.

At any time only one vibrator is in focus. You can set how long it is switched on for. Focus then goes to the next vibrator, and so on until a lap around the wrist has been completed. (It is in this sense that I mean the frequencies above, e.g. 5Hz with 3 vibrators delivers an individual pulse of vibration every 1/15th second.)

The next thing to do is to find a way of collecting statistics in as blind a way as possible. In addition to measuring the short-term impact on tremor I need to monitor the effect on bradykinesia and any long-term effects.

If anyone wants any help in building one of these devices, please get in touch.

John

I've heard that train rides are very soothing to PWP.

John,

I did hear back from the designer, who said more details will be published in 2017.

The original protope used six motors. You can see them inside the watch band like device. The actuators firing pattern is shown in this screen shot.

Pattern is 2000ms (2 seconds). Each state is for 100mS. Simple job for a micro. I was looking at using two "N" sized batteries (shorter than AAA) and 3D printing a housing with battery, microprocessor, and the six motors embedded into the band.

Haven't had any free time to work on this the past few weeks.

1 1 1 0 0 1 State One 0x25

0 1 0 0 1 0 State Two 0x12

0 0 1 0 0 1 State Three 0x05

1 1 1 0 0 1 State Four 0x35

0 0 0 1 1 1 State Five 0x07

1 0 1 0 0 0 State Six 0x24

0 1 1 0 0 0 State Seven 0x14

0 1 0 1 0 1 State Eight 0x15

1 0 0 0 0 0 State Nine 0x20

0 1 1 1 1 1 State Ten 0x1F

1 0 0 1 0 0 State Eleven 0x24

0 0 1 0 0 1 State Twelve 0x09

1 1 0 1 0 0 State Thirteen 0x34

0 0 1 1 1 1 State Fourteen 0x1F

1 0 0 1 0 0 State Fifteen 0x24

0 1 0 0 0 1 State Sixteen 0x11

1 0 1 1 1 1 State Seventeen 0x2F

0 1 0 0 1 0 State Eighteen 0x12

0 0 0 0 0 1 State Nineteen 0x01

1 0 0 0 0 0 State Twenety 0x20

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 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.