Mirror movements – A simple algorithm for mirror activity signal processing and normative values

• Published in: Neuroscience letters Vol. 803; p. 137186
• Main Authors: Castro, José, Pedrosa, Tomás, de Castro, Isabel, Swash, Michael, de Carvalho, Mamede
• Format: Journal Article
• Language: English
• Published: Ireland Elsevier B.V 23.04.2023; Elsevier
• Subjects: Adult; Aged; Aged, 80 and over; Algorithms; Electromyography - methods; EMG; Female; Healthy Volunteers; Humans; Isometric Contraction - physiology; Male; Middle Aged; Mirror activity; Mirror movements; Movement - physiology; Muscle, Skeletal - physiology; Normative values; Reference Values; Reproducibility of Results; Signal processing; Young Adult; Signal processing; Normative values; Mirror activity; Mirror movements; Emg
• DOI: 10.1016/j.neulet.2023.137186
• ISSN: 0304-3940; 1872-7972

•Mirror activity is a known phenomenon in some neurological diseases and in healthy subjects.•We propose a quick and reproducible method for analyzing and quantifying this activity.•Algorithm performance is on par with human operators.•Mirror activity is higher in the dominant side muscles.•We propose a set of normative values for clinical practice use. Mirror activity is an involuntary activation of a muscle when the respective contralateral muscle is contracting. This phenomenon has been described primarily in children and in disease states, and, more recently, also in healthy adults. Different ways of assessing mirror activity have been described. In this work we propose a simple protocol for quantifying the amount of mirror activity during a brief isolated full force isometric contraction of a given muscle. The signal was analyzed by a custom-built algorithm that detects the beginning and the end of muscle contraction. The amount of EMG signal on the mirror muscle in relation to the amount of EMG signal of the active muscle is then calculated. We studied 57 right-handed healthy subjects. Mirror activity was evaluated in the Abductor digiti minimi (ADM) and Tibialis anterior (TA) muscles during a 2–3 s full force isometric contraction. The intensity of mirror movement was represented as a percentage of the signal from maximal voluntary contraction. The performance of the algorithm for the detection of the beginning of muscle contraction was very good, when compared to 2 human operators. Intraclass correlation coefficient was excellent (0.998). The Bland-Altman plots showed similar performances of the algorithm and the human operators. We found a significant correlation of mirror activity with intensity and age. There was significantly more intense mirror activity in the left limbs (non-dominant) when compared to the right limbs. The upper limits of normality for mirror EMG signal was 27.4% for right ADM, 15.4% for left ADM, 10.4% for right TA and 2.1% for left TA. This simple protocol allows for an objective measurement of the amount of mirror activity. We propose this technique for investigation of neurological disorders.