Fully automated F-wave corridor extraction and analysis algorithm for F-wave analyses and MUNE studies

Figures

• 

  Figure 1. Electrode placements for median (left) and ulnar (right) nerve recording. R recording electrode, S stimulator, G ground.

• 

  Figure 2. An entire trace from the median nerve recording of a healthy control (stimulus artifact was discarded) and the result of noise reduction detail of the F-wave belonging to the same trace.

• 

  Figure 3. A trace from the median nerve recording of a healthy control before and after discarding stimulus artifact.

• 

  Figure 4. The signal of sum of all traces for a healthy control from a median nerve recording.

• 

  Figure 5. The first and the second peak of the autocorrelation function and the calculated distance between the locations of the global maximum and the second peak at the autocorrelation function for a healthy control from a median nerve recording [triangles: peaks] (left side of the red line was constructed by breaking the ordinate to observe the peaks clearly).

• 

  Figure 6. The maximum amplitude location searching window, approximate location for maximum amplitude location of F-waves which is acquired from autocorrelation function and the exact maximum amplitude location for the F-waves belonging to a healthy control recording [triangles: peaks].

• 

  Figure 7. The selected peak location (X = 118.55 ms) from the autocorrelation function and the maximum amplitude location of the F-wave (X = 23.35 ms) for a healthy control from a median nerve recording.

• 

  Figure 8. The extracted F-wave corridor from median nerve recording of a healthy control.

• 

  Figure 9. Linear line (estimated baseline) belonging to the original trace (blue colored) of a healthy control recording for the baseline correction and final result (purple colored) for that trace after the baseline correction.

• 

  Figure 10. Flow chart of both the automated F-wave extraction algorithm and baseline correction part.

• 

  Figure 11. Flow chart of the feature extraction algorithm from F-waves.

• 

  Figure 12. The most similar repeater F-wave pair belonging to a healthy control recording [lower similarity coefficient means more similar F-wave pairs].