1. Automated analysis of F-waves.
- Author
-
Artuğ, Tuğrul, Şirin, Nermin Görkem, Baslo, Mehmet Barış, Öge, Ali Emre, and Göker, İmran
- Subjects
- *
ELECTROMYOGRAPHY , *MOTOR unit , *NEURODEGENERATION - Abstract
F-waves are elicited in response to the supramaximal stimulus and they are consisted of the action potentials that are bounce back to muscle after invading anterior horn cell. F-waves' morphology and latency are variable. Morphology of the F-wave for a single motor unit is constant. F-waves with same morphology are known as repeaters. They can be used to estimate the number of motor units (MUNE). Signals are filtered from noise and first 2 ms part (stimulus artifact) is discarded. Maximum amplitude of Mresponse is determined for each signal. Mean value is calculated (MGloMax). F-waves are cut from both ends. Maximum (Fmax) and minimum (Fmin) amplitudes are calculated with locations. Signals with peak to peak amplitudes greater than 40μV are approved as F-waves. If a signal does not go down 40μV to the left and right in 3 ms from the Fmax location, it is floored to 0. All signals are aligned according to Fmax. Signal pairs closer than 0.5ms are determined as repeater F-wave candidates. If Fmax difference is lower than 10% and individual power difference is lower than 20% between the candidates, they keep their candidacy. Pairs having correlation coefficient more than 0.9 were also kept. A "similarity coefficient" is calculated relying from amplitude and power difference. If similarity coefficient is lower than 0.6 (determined by authors) than it becomes a repeater F-wave. Same repeater Fwaves are combined in the same basket. The mean of peak to peak amplitude (sMUP) for all signals are calculated. MUNE is calculated by dividing the MGloMax with mean sMUP amplitude. The number of F-waves, the number of repeater F-waves and the MUNE value can be seen on monitor. The most similar signal pairs can be plotted with similarity coefficient. [ABSTRACT FROM AUTHOR]
- Published
- 2018