Spatial distribution of surface EMG on trapezius and lumbar muscles of violin and cello players in single note playing

Musicians activate their muscles in different patterns, depending on their posture, the instrument being played, and their experience level. Bipolar surface electrodes have been used in the past to monitor such activity, but this method is highly sensitive to the location of the electrode pair. In this work, the spatial distribution of surface EMG (sEMG) of the right trapezius and right and left erector spinae muscles were studied in 16 violin players and 11 cello players. Musicians played their instrument one string at a time in sitting position with/without backrest support. A 64 sEMG electrode (16 x 4) grid, 10 mm inter-electrode distance (IED), was placed over the middle and lower trapezius (MT and LT) of the bowing arm. Two 16 x 2 electrode grids (IED = 10 mm) were placed on the left and right erector spinae muscles. Subjects played each of the four strings of the instrument either in large (1 bow/s) or detache tip/tail (8 bows/s) bowing in two sessions (two days). In each of two days, measurements were repeated after half an hour of exercise to see the effect of exercise on the muscle activity and signal stability. A "muscle activity index" (MAI) was defined as the spatial average of the segmented active region of the RMS map. Spatial maps were automatically segmented using the watershed algorithm and thresholding. Results showed that, for violin players, sliding the bow upward from the tip toward the tail results in a higher MAI for the trapezius muscle than a downward bow. On the contrary, in cello players, higher MAI is produced in the tail to tip movement. For both instruments, an increasing MAI in the trapezius was observed as the string position became increasingly lateral, from string 1 (most medial) toward string 4 (most lateral). Half an hour of performance did not cause significant differences between the signal quality and the MAI values measured before and after the exercise. The MAI of the left and right erector spinae was smaller in the case of backrest support, especially for violin players. Back muscles of violin and cello players were activated asymmetrically, specifically in fast movements (detache tip/tail). These findings demonstrate the sensitivity and stability of the technique and justify more extensive investigation following this proof of concept. (C) 2016 Elsevier Ltd. All rights reserved.