Program

CO38-007

Oral Communication

Proposal of tremor quantification method

Dr Marina GUIHARDa, Dr Emilie HUTINa, Dr Inke Marie ALBERTSENa, Dr Valentina MARDALEa, Prof Jean-Michel GRACIESa

a Laboratoire Analyse et Restauration du Mouvement, EA BIOTN, Hôpitaux Universitaires Henri Mondor, Assistance Publique - Hôpitaux de Paris, Université Paris-Est Créteil (UPEC), France

Introduction: The diagnosis and the follow-up of pathological tremor (rest, essential or cerebellar) are difficult, especially in early stage.1 This study defines tremor quantification indexes from a spectral analysis of EMG signals.

Methods: Bilateral EMG of the wrist flexors/pronators (FlPr) and extensors/supinators (ExSu) were recorded (Fe=1000Hz) in one patient (50y) with a left side tremor clinically observed. Eight conditions were tested during 10s: at rest without any distraction (1), with cognitive distraction (2), with contralateral motor task (3), posture with the wrist pronated (4), semi-pronated (5), supinated(6), flexed elbows with proned wrist close to the thorax (7), walking (8). The EMG signal envelop is obtained by a band-pass [20;450Hz] and a low-pass (30Hz) filters. A power spectral density study (DSP) from 1.5 to 12Hz leaded to the calculation of the main frequency, the maximum DSP and the rhythmicity index (IR=area under the main frequency peak/total area).2

Results: The main frequency is similar in all conditions (FlPr G 5.4±0.1Hz, ExSu G 5.6±0.3Hz ; FlPr D 7.2±0.3Hz, ExSu D 7.4±0.4Hz, mean±standard deviation). The maximum DSP is higher on left side independently to the task (eg. Rest, ExSu G 42.33mV2/Hz, D 0.45mV2/Hz). The IR is also higher on left side with a reduced difference with cognitive or motor distraction (muscles averaged, tasks 4-8, G 47±13%, D 27±13%; tasks 2-3, G 49±16%, D 39±20%). The maximum DSP and the IR are higher in extensors than in flexors in each task (postures 4-7, FlPr G 38±18%, ExSu G 43±17%).

Conclusion: This method of tremor quantification, based on the spectral density analysis of EMG signals, shows invariant tremor main frequency, whereas the maximum DSP and the IR depend on the muscle. Only the IR seems to be impacted by the task (without vs with distraction) and also could be used in the follow-up of tremor patient.

References

1Elble et al. Factors influencing the amplitude and frequency of essential tremor. Movement Disorders 1994:9;589-96.

2Velickovic, Gracies. Keys to identifying and treating tremor. Geriatrics 2002:57;32-6.

Keywords : Tremor, Rhythmicity, EMG, Power spectral density.