Objective measures of fatigue are similar to objective measures of pain - very difficult. Blood chemistry may provide a measure of enzymes from muscle break down. Change in metabolic rate measured by heart rate, respiration and heat production may also be an indicator. Including fatigue after hard work vs. fatigue after long periods without sleep may be very difficult.

Maybe a combination of metabolic indicators and alertness? I'll be very interested in folks comments.

Hi,

In addition to the mentioned methods by James and Javid, surface EMG signals can also be beneficial if you are willing to determine physiological fatigue of a special muscle. Moreover, by extracting some features including RMS (root mean square), MNF (mean frequency), MDF (median frequency), AIF (averaged instantaneous frequency), some wavelet features, … you can quantify fatigue and indicate its manifestation during a time period.

Kind regards,

Fariba

Is it possible to detect the fatigue level of user based on their hand's motion? For example, the fatigue level of an employee can be detected based on their mouse's motion.

It seems to me that I've seen references to EMG/accelerometer approaches to measure muscle tremor. I think that the difficulty would be establishing testing and training data which defines a level of fatigue under a variety of conditions. Is fatigue measured by alertness, endurance or deft execution of small physical tasks (fine motor tests), concentration and cognition accomplishing small mental exercises, or a subjective survey?

The interesting research project might study the the relationship amoung those measures. A objective measure of fatigue level may be a combination of measures: HRV, heart rate, gross EEG measures and EMG/tremor. Measured repeatedly over time, with exercise and sleep deprivation, you'll see a good calibration set. My guess would be that across a range of subjects the "good" measures will be vary widely.

For smooth muscles, check this out:

WANG Gang, REN Xiao-mei, LI Lei, WANG Zhi-zhong

Multifractal analysis of surface EMG signals for assessing

muscle fatigue during static contractions*

Journal of Zhejiang University SCIENCE A

IISSN 1673-565X (Print); ISSN 1862-1775 (Online)

It worked with us for some preliminary EMG taken while arm-wrestling...

A computer mouse is a very poor measurement device, when it comes to small amplitude signals like tremor, even jerk is filtered out... and the arm and its big muscles are supported and thus less prone to fatigue. Either you use a digitizing tablet and an appropriate pen (then you will induce fatigue on purpose) or you use EMG recordings on the lower forearm. What about keystroke speed on a keyboard? Reliable time measurement with sufficient resolution is not easy there...

An excellent review paper:

https://skydrive.live.com/redir?resid=300CC2296DC3BCF2!190

Dear Anh Nguyen, the answer is simply enough. Just take the surface EMG of the muscles involved in the movement and measure the Mean Frequency (MNF) or the Median Frequency (MDF). If you need more details, please send me a specifical request to my E-mail.