Adaptive and personalized neuromotor rehabilitation of persons with Multiple Sclerosis: from characterization to exploitation of residual sensorimotor abilities using a body-machine interface

Project Information

REMAp

Grant agreement ID: 896892

DOI

10.3030/896892

Project closed

EC signature date 14 April 2020

Start date 4 January 2021

End date 3 January 2023

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 171 473,28

EU contribution

€ 171 473,28

Coordinated by

UNIVERSITA DEGLI STUDI DI GENOVA
Italy

Periodic Reporting for period 1 - REMAp (Adaptive and personalized neuromotor rehabilitation of persons with Multiple Sclerosis: from characterization to exploitation of residual sensorimotor abilities using a body-machine interface)

Reporting period: 2021-01-04 to 2023-01-03

Summary of the context and overall objectives of the project

Multiple sclerosis (MS) is one of the world’s most common neurologic disorders. It is a chronic progressive disease that leads to deficits of motor, sensory and cognitive functions. Rehabilitation is a key element for people with Multiple Sclerosis (PwMS) to mantain and maximize their motor functions, and technology-aided neurorehabilitation approaches have been proved to overcome some limits of traditional treatments. In particular, body-machine interfaces (BoMIs) seem very promising. With BoMIs it is possible to exploite the abundance of muscles and degrees of freedom of the motor apparatus, providing for a high level of dexterity that even with a progressive neurologic disease like MS is not totally lost, but is still enough to be used by the BoMI to empower the user to interact with the environment controlling external devices in an efficient way.

The problems addressed in REMAp are: 1) lack of precise and sensitive tools to evaluate and treat upper limb sensorimotor impairment after MS; 2) BoMI never been tested with PwMS; 3) no studies investigated neural correlates of motor training in MS in a complete manner, from muscle to brain; 4) lack of evidence of modifications in proprioception after BoMI training.

Filling these gaps is important for society because it will advance the state of the art helping better understanding the pathophysiology of the disease and to advance the state of the art of BoMIs and rehabilitation after MS.

Overall objectives:
O1 - To design and implement a BoMI that integrates muscle activity and motion signals for personalized motor training of people with MS
O2 - To identify and understand changes in the organization of the central and peripheral nervous system in response to BoMI rehabilitative interventions in MS
O3 - To assess if a BoMI training enhancing the exploration of new movement coordination patterns has a positive impact on proprioception

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

O1:
the technical design (software and hardware) of the BoMI included i) correct integration of EMG and IMUs signals, ii) study and design of the algorithms for the principal component and autoencoder used for dimensionality reduction; iii) programming of the task to be performed with the BoMI and relative evaluation protocol.
Ethical approval was obtained for both populations involved in the study.
The results of the experiment on healthy subjects allowed us to chose the linear algorithm based on principal component as the best algorithm to be used for the training of PwMS. Such training proved the hybrid BoMI to be an effective tool to exercize upper limb coordination. The training was well tollerated by PwMS that also improved in the clinical scales (FuglMeyer test for upper limbs, 9Hole peg tes).

O2:
Imaging data were acquired with a 3.0 Tesla Siemens Prisma scanner, equipped with a 64-channel head and neck coil. The MR imaging protocol included structural, diffusion DTI and resting state fMRI sequences for brain and spine. Results show that after BoMI training as expected PwMS did not report changes in lesion load. From DTI analysis there was an increase in FA of the corpus callosum that was correlated to an improvement of the clinical scale 9HPT. Such thing is also in line with the improvement in the proprioceptive task (O3). Additionally after training there was an increase of FC between the thalami and both cerebellar hemispheres and an increase of FC between both cerebellar hemisphere and the insula and superior temporal gyrus. These two networks are networks involving cortical areas responsible for sensory integration, motor modulation and execution. The fact that both hemispheres reported modifications is in line with the fact that the BoMI training was a bilateral training involving both upper limbs and therefore left and right brain hemispheres reported some modifications related also to improvements in the clinical tests (9HPG and FMUE).

O3:
A new proprioceptive task to assess upper limb three dimensional sense of proprioception by using an immersive virtual reality environment was developed and tested. It is implemented in Unity and it requires the Oculus and two controllers. In the VR environment, subjects performed 3different matching taks: unilateral, bimanual with and without memory. Each task consisted in reaching a target presented in the 3-D virtual space with one hand and then move the same (if the task is unilateral) or the other hand (if the task is bilateral) in order to match the position of the first moved hand. This second movement was performed with the subjects blind-folded such that they can only rely on their sense of proprioception to execute the movement. Results indicated that this test was able to differentiate proprioceptive deficit of PMS with respect to their control (age and sex matched) subjects. Moreover proprioceptive deficit of PwMS after BoMI training decreased, meaning that the complex nature of the body signals used by the BoMI and the way they are combined to carry out the rehabilitative task, made users more aware of their body movements.

Dissemination results:
- I submitted 2works to the IEEE International Conference for Biomedical Robotics and Biomechatronics 2022.
- I organized a minisymposium at the IEEE/EMBS Conference on Neural Engineering on Body-Machine Interface titled: “Body-machine interface for empowering humans: from rehabilitation to robot control”
- I participated at the event 'Science is wonderful! 2021'
- I was invited twice by Scuola di Robotica in Genoa to give two webinars strictly connected to my research topic.
- I shared my research experience and knowledge on human machine interfaces I gained in this project to master students (2nd year) of Robotics and Bioengineering programs at the University of Genova.
- As a Maire Curie fellow I was invited in 2021 and 2022 by the European Research Office of the university to give a talk about my project and my experience during the Masterclass MSCA organized for researchers interested in applying to the MSCA individual fellowship program.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

This work resulted in the following main scientific and technological achievements:

- development of a new hybrid BoMI to assess and train motor abilities in people with MS.
- pipeline to characterize brain and spinal cord modifications after motor training and training-induced proprioceptive changes.
- system and tools to characterize proprioceptive deficit of upper limbs in a three dimensional environment using the Oculus.

These achievements contributed to improve the state of the art for the following reasons
1) a novel method and class of highly adaptive, personalized BoMIs that maximize functional movements restoration promoting functional reorganization of body abilities and multi-joint coordination lost to MS;
2) the identification and characterization of biomarkers of motor impairment at different levels of the nervous system - muscles, spine and brain
3) a quantitative understanding of how the increased dexterity after training and changes in cortical and spinal motor pathways will impact also sensory pathways reorganization.

These BoMIs employ a technology that can be rather inexpensive and therefore can be financially accessible to a broad body of potential users. It is noteworthy that companies as influent as Meta have increased their interest in wearable technologies for human-machine interfaces, suggesting that these devices have a bright future.

experimental set-up of the training session