Frequency analysis of a forearm/hand Essential Tremor

 Here is what my arm tremor looks like with a bit of frequency analysis applied to the raw data.

You can see that the data clusters around 5hz. There are a few data points in the 4.6 to 4.9 range and a larger spread on the upside from 5.1 up to 6.2 with a few scatter points all the way up to 7.0 hz.

I’ve not been privileged to see frequency plots of other people with Essential Tremor.  The few other charts I have seen show a similar profile.  

This particular chart was created with Google sheets.  Google is responsible for the somewhat odd numbering of the X-axis of this chart.  In an attempt to overcome this auto-numbering “feature” the data was reduced to a single decimal point but the chart failed to acknowledge this fact.

The data was captured this afternoon, Sept. 18, 2021 using the latest version of SensorGlove.ino.  The sample rate was 250hz and about 40 seconds of data was captured on an SD card.  I held my left arm in the “Wing Beating” position while recording this data set.

The data capture equipment consists of an Arduino Mega 2560, a Seeed SD card, and an Adafruit LSM6DS33 + LIS3MDL - 9 DoF IMU with Accelerometer/Gyroscope/Magnetometer. The IMU is mounted on a Velcro strip which positions it on the back of the hand.  The ‘X’ axis is aligned parallel to the wrist joint.  The ‘Y’ access is pointed in the direction of the fingers, and the ‘Z’ axis is normal to the hand.  The IMU, wrist band, and the electrical lead to the sensor have a combined weight of 0.8 oz.  As such they do not interfere with the free movement of the forearm/hand.

The raw data was post-processed into a CSV format file.  A low pass filter was then used on the raw data to remove most of the high frequency noise.  The tremor frequency was calculated by determining the distance between every second zero-line crossing point. With a sample rate of 250hz the average error rate should be 2% or less.  A plot of the data confirms that assumption.  What was surprising was the spread of the frequencies particularly on the upside of the 5hz peak. 

All this suggests that further work is needed in the frequency analysis for use as an input to a tremor stimulus device.  The measurement and analysis of tremors needs to be expanded beyond postural tremors.  The next prototype of the sensor glove should provide the ability to measure tremors across the 7 degrees of freedom of the arm.  Once we have the ability to measure and quantify this, the  Sensor Glove can be expanded to include thumb and finger tremors.

All fun and games for the Citizen Scientist. AMA

(Please note: This information should not be used as a substitute for medical treatment and advice. Always consult a medical professional about any health-related questions or concerns.)