Notes on Ninapro database, muscle synergies, MEMS

10 Apr 2022

1. Ninapro Database
2. Accelerometer, Gyroscope and Magnetometer sensors: what are they measuring?
3. Muscle Synergies in sEMG

Ninapro Database

There are 10 datasets in NinaPro. Since the goal of my project is to estimate finger position as opposed to grip/grasp. Database 8 seems very suitable for the task.
The paper describing this dataset can be found here. DB 8 as the documentation clearly suggest should be used for regression problem and not classification. “Therefore, the use of stimulus/restimulus vectors as target variables should be avoided”)

10 subjects able-bodied + 2 right-hand transradial amputees
3 datasets collected from each subject
- Dataset 1 & 2: 10 repetition/movement
- Dataset 3: 2 repetition/movement
Each dataset consists of
- 9 movements + rest
- movements performed bilaterally
- movement duration: 6-9 sec + 3 sec rest
- Each trial includes reaching the position (flexion) then extension
Can use datasets 1 & 2 for training and hyperparameter tuning, dataset 3 for testing, or merge 1 & 2 and do a cross-validation then test on 3.

Micorelectromechanical systems (MEMS) are components combining mechanical and electrical elements into a small devices (of size in the micrometer range).
Accelerometer sensor measures the acceleration (obviously 😀) ($m/s^2$). Recap high-school physics, according to Newton’s Second law of motion: the $acc \propto force$. It measures either static or dynamic acceleration. In static acceleration, there is a constant force acting on the object (eg. gravity, friction). Dynamic acceleration, the acting forces are non-uniform, for example, car crashing leading to acceleration.
Gyroscopes measure deflection from a given orientation and angular velocity. Said differently, they measure rotational motion (in $\circ/s$ or revolutions per second).
Magnetometer sensor measures relative changes in magnetic field at a particular location.

In voluntary movement, the central nervous system (CNS) has its own way to coordinate muscle activations. We do not still know for sure how this is happening but one thing that we do know is that the CNS must be doing some sort of dimensionality reduction. This is referring to the fact that muscle activities are coordinated, synchronized. So the hypothesis is the CNS might be “representing all useful muscle patterns as combinations of a small number of generators” d’Avella, A., Saltiel, P. & Bizzi, Nat Neurosci, 2003. in order to
allow for a wide range of degrees of freedom and great flexibility. In other words, there is only a small subset of controllers (i.e muscle synergies) that are combined in space and time to generate many muscle activity patterns.
People usually analyze such coherent activations of muscles using muscle synergies analysis. This refers to analyzing muscle patterns to extract the underlying “sources” for the observed muscle patterns. The generated muscle patterns are models are a “combinations of time-varying muscle synergies, that is, coordinated activations of a group of muscles with a specific time course for each muscle” d’Avella, A., Saltiel, P. & Bizzi, Nat Neurosci, 2003. .

Note: It seems a very similar idea to blind source separation for decomposing the sources from a mixed signal (more on that later 🙃)