G. Pregnolato1, D. Rimini1, F. Baldan1, L. Maistrello1, S. Salvalaggio1, A. Favetto2, N. Celadon3, P. Ariano2,3, A. Turolla1
1San Camillo IRCCS Hospital, Laboratory of Rehabilitative Technologies, Venice, Italy, 2Fondazione Istituto Italiano di Tecnologia, Artificial Physiology Group, Centre for Sustainable Future Technologies, Turin, Italy, 3Morecognition Srl, Turin, Italy
Background: Recovery of upper limb and hand function is fundamental for autonomy restoration after stroke. Innovative technologies represent a valid support for the delivery of rehabilitation treatments. Embedding surface electromyographic (sEMG) into wearable devices, allows the customisation of rehabilitation exercises, based on the functional profile of each patient.
Purpose: Clinical stratification of stroke survivors to detect their ability to control a sEMG wearable device targeted to hand rehabilitation.
Methods: Patients with first stroke, ischaemic or haemorrhagic, were enrolled in a cross-sectional study. They were clinically assessed, then tested on the ability to control the sEMG device. The device is a wearable armband composed by 8 bipolar electrodes used to classify muscle pattern activation at patients’ paretic forearm. The test is composed of 10 hand and fingers functional tasks to be performed with the paretic hand. Baseline and activation sEMG signals were recorded and compared for each movement. Three conditions representing absent, partial or full control of the device were defined and logistic regression models were used to identify clinical features describing the group each patient belongs to. Clinical cut-off for each strata was identified by odds ratio.
Results: 117 stroke survivors were enrolled (no control, N = 19; partial control, N = 21; full control , N = 76). The logistic regression models indicated that Fugl-Meyer Upper Extremity (FM-UE) ≥ 10/66 (OR=1.15, CI95%: 1.18-1.98) and FM sensation ≥ 43/48 (OR=1.22, CI95%: 1.04-1.43) are needed to perform 1 to 4 tasks (AUC = 0.95). Then, FM-UE ≥ 18/66 (OR=1.65, CI95%: 1.22-2.22) and Modified Ashworth Scale score of 0/4 at Flexor Carpii muscle (OR=0.45, CI95%: 0.26-0.92) are needed to perform 5 to 9 tasks (AUC = 0.97). Finally, FM-UE ≥ 18/66 (OR=1.11, CI95%: 1.07-1.15), is needed to control all 10 tasks (AUC = 0.91).
Conclusion(s): Patients stratification in rehabilitation is fundamental to address patients specific needs. Those results suggest that motor, sensation and muscle tone functions are critical clinical features to control sEMG devices for hand rehabilitation.
Implications: A new concept for therapy individualization, based on patient clinical stratification, allows the best referral to therapy.
Funding, acknowledgements: None.
Keywords: stroke rehabilitation, hand recovery, wearable technology
Topic: Innovative technology: robotics
Did this work require ethics approval? Yes
Institution: San Camillo IRCCS Hospital, Venezia, Italy
Committee: Ethics Committee for Experimentation of ULSS12 and San Camillo IRCCS Hospital
Ethics number: CE cod. 2016.29 – More
All authors, affiliations and abstracts have been published as submitted.
