Design and Control of an Actuated Thumb Exoskeleton for Hand Rehabilitation Following Stroke

Pictures at the URL. Anything is helpful to get our hands back. I could use thumb help since my thumb does not spread 90 degrees to my first finger.
http://robolab.cse.unsw.edu.au/conferences/ICRA-2011/data/papers/1170.pdf
Abstract— Chronic hand impairment is common following
stroke. This paper presents an actuated thumb exoskeleton
(ATX) to facilitate research in hand rehabilitation therapy. The
ATX presented in this work permits independent bi-directional
actuation in each of the 5 degrees-of-freedom (DOF) of the
thumb using a mechanism that has 5 active DOF and 3 passive
DOF. The ATX is able to provide considerable joint torques for
the user while still allowing backdrivability through flexible
shaft transmission. A prototype has been built and experiments
were conducted to evaluate the closed-loop position control.
Further improvement and future work are discussed.
I. INTRODUCTION
Hand impairment is a prevalent outcome for a variety of
neuromuscular disorders, such as stroke. Up to 795,000
people in the U.S. experience a stroke each year [1]. Of these,
60-75% will live beyond one year after the incidence,
resulting in a current stroke population of 6.5 million [1]-[2].
Arm function is acutely impaired in a large majority of those
diagnosed with stroke [3]-[5]. Furthermore, acute
hemiparesis presages chronic hemiparesis in over 40% of
individuals [3], [4]. Chronic deficits are prevalent in the distal
upper extremities, especially with regard to finger extension
[6].
This distal limb impairment can be especially disabling, as
proper hand function is crucial to manual exploration and
manipulation of the environment. Indeed, loss of hand
function due to neuromuscular disorders frequently prevents
effective self-care and limits employment opportunities. One
study reported that more than half of the subjects they
observed were dependent on others for help in the activities of
daily living six months post-stroke [7].
An assortment of interventions has been tried in an effort to
improve function or to treat the resulting peripheral
alterations following stroke. Those with the most success to
date tend to focus on repetitive practice. Indeed numerous
studies employing the constraint-induced technique, in which
focus is placed on intensive practice with the impaired arm
without using the less impaired arm, have shown
improvement in hand capabilities [8]-[9]. This supports the
observations in animal models of stroke in which practice
Furui Wang, Milind Shastri, Vikash Gupta and Nilanjan Sarkar are with
the Department of Mechanical Engineering, Vanderbilt University, Nashville,
TN, 37212, USA (email: {furui.wang, vikash.gupta, milind.shastri,
nilanjan.arkar}@Vanderbilt.edu).
Christopher L. Jones, Christian Osswald and Xuan Kang are with the
Department of Biomedical Engineering, Illinois Institute of Technology,
Chicago, IL, 60616, USA (email: jonechr@Iit.edu, cosswald@iit.edu,
xuankang@hotmail.com,).
Derek G. Kamper is with the Department of Biomedical Engineering,
Illinois Institute of Technology and the Rehabilitation Institute of Chicago,
Chicago, IL, 60611, USA (e-mail: d-kamper@northwestern.edu).
appears to be the primary factor leading to synaptogenesis
and brain plasticity [10]-[11].
Unfortunately, many stroke survivors do not possess
sufficient sensorimotor control to practice the desired
movements. For the upper extremity, robots have been
created to assist with therapeutic training of the wrist, arm and
shoulder [12]-[15]. It has been reported that robot-delivered
sensorimotor training enhanced the motor performance of the
exercised shoulder and elbow with improved functional
outcome [16] and that practicing with a robot that assisted
reaching movements helped the users learn how to generate
smoother unaided reaching trajectories [17].
In recent years, a number of devices have been developed
expressly for or applied to hand rehabilitation. These include
both commercial products, such as CyberGrasp (Immersion
Corporation, San Jose, CA), the Hand Mentor (Kinetic
Muscles Inc., Tempe, AZ) and the Amadeo System
(Tyromotion GmbH, Graz, Austria), and experimental
devices, such as Rutgers Master II-ND, HWARD and
HandCARE, among others [18]-[20].
Questions remain, however, as to how best to use robotic
devices to facilitate rehabilitation. Should the device assist or
resist movement? Should movement error actually be
augmented, as some have suggested [21]? Should emphasis
be placed on practice of movement of individual joints [22] or
on the coordination of multiple joints?
The majority of the systems developed for the hand do not
allow for independent control of each DOF of the
finger/thumb joints, making it difficult to answer these
questions. This is especially true for the thumb, which is
typically modeled with five degrees-of-freedom (DOF)
[23]-[24]. An 18 DOF hand device has been developed in
Japan for hand and wrist rehabilitation following stroke [20].
This device actuates 4 DOF for the thumb. The joint torques
that it can provide, however, are limited to roughly 0.3 N-m, a
value that may not be sufficient for resistive or perturbation
therapy paradigms.
Thus, in this work, we are designing an actuated thumb
exoskeleton (ATX) with 5 active DOF that allows
independent actuation of each DOF of the thumb. The ATX is
able to provide considerable torque to overcome the
excessive coactivation or increased stiffness in the affected
thumb. We present our first prototype of the ATX and initial
experimental results in this paper. The paper is organized as
follows: Section II describes the mechanical design of the
ATX; Section III presents the kinematic analysis; Section IV
introduces the real-time control system; Section V shows the
instrumentation and initial experimental results; and Section
VI discusses the necessary improvement and Section VII
concludes the paper and proposes future work.