Stroke is an acute focal injury in the brain which results in functional deficits. Among them, deficits in motor control entail high impact on quality of life. Improving upper limb functioning through neurorehabilitation is a major therapeutic target in stroke rehabilitation to maximize patients' functional recovery and reduce long-term disability. Motor deficits such as muscle weakness and spasticity are commonly assessed by means of clinical scales. However, reliability and responsiveness issues limit the ability of these scales to monitor the motor recovery over time. In fact, clinical scales are often supported by objective measures derived from electromyography (EMG) or kinematics - which quantify motor actions from a functional point of view - or from electroencephalography (EEG) - to quantify motor impairment and recovery of the central nervous system.
Aim of the project is to develop a unified multimodal framework integrating EEG, EMG and kinematic signals, able to effectively evaluate the quality of upper limb movements in stroke patients. Such a tool would provide a set of multimodal indices describing and quantifying the upper limb motor impairment induced by the stroke and assessing the recovery induced by a motor rehabilitative intervention.
In the three years of the project, we plan to: i) characterize the motor impairment in stroke patients by means of multimodal features extracted from EEG, EMG and kinematics data; : ii) validate the proposed framework on a dataset collected from 20 healthy subjects and 20 stroke patients during the execution of simple and complex motor tasks; iii) use the measures of motor impairment to assess the recovery induced by a motor rehabilitation treatment supported by brain-computer interface in 20 stroke patients.
The success of the project will provide clinicians with an instrument aiding the planning of neurorehabilitation interventions and the timely monitoring of outcomes.
This research project takes place in the context of motor impairment evaluation and the assessment of rehabilitation strategy supporting post-stroke upper limb motor recovery. The multimodal framework proposed in the project would allow to characterize the motor impairment induced by the stroke and the related rehabilitation considering all the parties involved as a unique system. To the best of our knowledge this is the first time in which the upper limb motor action is studied in physiological and pathological conditions from a global perspective including the central command and the peripheral output. In fact, motor impairment could be originally attributed to dysfunctions in brain activity due to damages produced by the stroke and then such impairments could have a high impact on the functionality of the affected arm. EEG signals will describe the modifications in electrical activity provoked by the stroke and will monitor the plasticity phenomena related to the motor rehabilitation. EMG signals usually give information about the onset and intensity of the muscle activations. But that information does not complete the evaluation of the motor abilities. Therefore, we propose a newly integrated approach in which the motor abilities will be assessed from both myoelectric signals and mechanical magnitudes. Forces involved in the flexion and extension of the fingers and joint angles could build up the knowledge about upper limb movements in stroke patients.
The development of such multimodal tool for the description of motor impairment in stroke and the assessment of motor recovery will impact on different aspects of the stroke rehabilitation:
1) the use of multimodal features for the identification of motor impairment in stroke patients would provide the clinicians with a complete description of the movement to be used for planning the rehabilitation treatment (identification of voluntary movements to be reinforced and pathological synergies, co-contraction of antagonist muscles to be discouraged for mitigating spasticity).
2) In an exploratory manner, the framework developed in the project will be used on its own for the assessment of a rehabilitative intervention supported by BCI. Once the parameters have been defined to evaluate motor impairment, the same principle could be applied in the evaluation of efficacy of other rehabilitative treatments for the upper limb. This tool will be a user-adjustable system, which could follow the subject's improvements and changes in time.
3) The identification of indices characterizing the motor impairment and its recovery will provide information to be used in the construction of new BCIs based on multimodal features to be trained in real time. Such BCIs will rehabilitate the entire motor action from the central system to the periphery.
References
---------------
1. Watkins et al., Clin. Rehabil., 2002.
2. Pollock et al., Cochrane Database Syst. Rev., 2014.
3. Coscia et al., Brain J. Neurol., 2019.
4. Pichiorri et al., Ann. Neurol., 2015.
5. Pandian and Arya, J. Bodyw. Mov. Ther., 2014.
6. Vinti et al., Off. J. Int. Fed. Clin. Neurophysiol., 2013.
7. Valevicius et al., J. Electromyogr. Kinesiol., 2018.
8. Hebert et al., J. Rehabil. Res. Dev., 2014.
9. Alt Murphy et al., J. Neuroengineering Rehabil., 2006.
10. Alt Murphy et al., Neurorehabil. Neural Repair, 2013.
11. Grefkes and Fink, Lancet Neurol., 2014.
12. Grefkes et al., Ann Neurol, 2008.
13. Mathiowetz et al., Am. J. Occup. Ther., 1985.