*3.4. EMG Noise Results*

High amounts of noise were very common for patients in the ICU. With careful selection of electrode placement, however, it was usually possible to optimize noise sufficiently to allow for motor mapping (Figure 3). Average noise amplitude after optimization was 64 ± 58 µV (median noise amplitude: 43.5 µV). Trial-and-error approaches were often required for successful optimization, which entailed up to 10 different electrode placement trials per muscle. For each muscle group, particular anatomic locations for electrode placement (muscle reference electrode and grounding electrode) emerged. Despite identifying these sites as having a higher probability for noise mitigation, optimization via trial and error was nevertheless required in most cases (Figure 4). In many cases, noise optimization was the most time-consuming part of the procedure.

**Figure 3.** Electromyography (EMG) noise optimization. This figure illustrates the effect of electrode placement on noise level. All placement spots are common for neutral electrode placement, yet significant differences can be observed. At bedside, optimal positioning often requires testing of different spots until an adequate noise level is achieved. EMG scales are equal within the same row.

**Figure 4.** Placement spots for neutral electrodes. This figure illustrates anatomic landmarks that often emerged as viable for neutral electrode placement. 1: medial and lateral side of thumb, interphalangeal joint; 2: tendon of biceps brachii muscle; 3: medial epicondyle of humerus; 4: acromion; 5: tendons of hand flexors; 6: tendons of foot extensors; 7: medial malleolus of tibia; 8: lateral malleolus of fibula; 9: medial epicondyle of femur.
