Assessments are totally fucking worthless, we need EXACT STROKE PROTOCOLS that provide 100% recovery. WHEN THE HELL WILL YOU GET THERE? When you are the 1 in 4 per WHO that has a stroke will you be satisfied with this crapola and not getting recovered?
Assessment of Motor CompensationPatterns in Stroke Rehabilitation Exercises
Ana Rita Cóias
ana.coias@tecnico.ulisboa.pt
Alexandre Bernardino
alex@isr.tecnico.ulisboa.pt
Institute for Systems and Robotics
Instituto Superior Técnico, ULisboa
Lisboa, PT
Abstract
The increasing demand concerning stroke rehabilitation and in-home exercise promotion requires objective methods to assess patients’ quality of
movement, allowing progress tracking and promoting consensus among
treatment regimens. In this work, we propose a method to detect diverse
compensation patterns during exercise performance with 2D pose data to
automate rehabilitation programs monitorization in any device with a 2D
camera, such as tablets, smartphones, or robotic assistants.
1 Introduction
With the escalating demands towards stroke rehabilitation and the increase of in-home exercise recommendations [2], the need for new means
to evaluate patients’ motor performance has risen [4, 7]. In conventional
assessment tests, therapists assess movement quality based on observation, thus being highly subjective [4]; with the degree of experience implying distinct treatment approaches [7]. Quantitative and objective methods allow patients’ progress tracking, impaired movements’ understanding, and formulation of standard therapy regimens [4, 6].
Patients’ physically impaired often exhibit compensation behaviors to
accomplish a task. Motor compensation is the presence of new movement
patterns derived from the adaptation or substitution of old ones, which
might help patients’ execute a task [5]. New patterns can include the use
and activation of additional or new body joints and muscles. Most typical
compensation behaviors are trunk displacements, rotation, and shoulder
elevation. These functional strategies are commonly observed in reaching
and are highly related to severe impairment levels [5].
Early on the recovery process, the use of compensation strategies promotes patients’ upper limb participation in task performance. However,
their persistence may obstruct real motor function recovery and must be
reduced during therapy through appropriate exercise instructions [5].
In this work, we present a method to assess quantitatively motor compensation from video frames during upper limb exercise performance. We
have created a label set (Table 2) for each video frame of the dataset regarding the observed compensation patterns. We then explore two methods to assess these patterns based on 2D pose data enabling this kind of
analysis with widely available RGB cameras.
No comments:
Post a Comment