Yu-Kyong Choe

Multidisciplinary stroke rehabilitation delivered by a humanoid robot: Interaction between speech and physical therapies

Background: A great number of stroke patients pursue rehabilitation services in multiple domains (e.g., speech, physical, occupational). Although multidisciplinary and interdisciplinary approaches to stroke rehabilitation are considered desirable, it is largely unknown how the intervention in one domain affects the progress in others. Aims: The current study investigated the interaction between speech therapy and physical therapy. Additionally, the feasibility of utilising a humanoid robot in stroke rehabilitation was described. Methods & Procedures: A 72-year-old male chronically challenged by aphasia and hemiparesis completed speech and physical therapy tasks in the sole condition (Speech Only, Physical Only) and in the sequential condition (Speech & Physical). The therapy activities were delivered by a humanoid robot. Outcomes & Results: Greater gains in speech and physical functions were obtained during the sole condition than in the sequential condition, suggesting a competitive interaction between speech and physical therapies. Conclusions: The cross-domain competition can be accounted for by fatigue, participant characteristics, and task characteristics. Objective data on speech and physical functions and subjective data on perceived quality of life indicate positive outcomes in this single case. These findings warrant further research on the feasibility and utility of humanoid robots in stroke rehabilitation.

Towards extended virtual presence of the therapist in stroke rehabilitation

This paper considers the use of humanoid robots in residential stroke care to facilitate both direct and indirect interaction between clients and therapists. Direct interaction is realized through a humanoid-mediated teletherapy where a therapist assesses the motor function of a patient and provides therapy customized to the individual. During the teletherapy sessions, the therapist uses a simple speech interface to program therapeutic behavior and activity. Indirect interaction is implemented by the therapist-programmed artifact where a humanoid robot delivers therapeutic activities to the stroke patient in the absence of the therapist. We propose that such an approach can amplify the outcome per hour of therapist time. Outcome data from the current study indicate that the therapist can successfully provide customized therapy to individuals in residential settings and warrant further study.

Upper-limb exercises for stroke patients through the direct engagement of an embodied agent

In this case study, we examine the functional utility of an embodied agent as an interactive medium in stroke rehab. A set of physical rehab exercises is conducted through the direct engagement of an embodied agent, the uBot-5. Based on the preliminary data, we argue that a general-purpose embodied agent has a potential to functionally complement human therapists in providing rehab to stroke patients.

Evidence of cue use and performance differences in deciphering dysarthric speech.

There is substantial performance variability among listeners who transcribe degraded speech. Error patterns from 88 listeners who transcribed dysarthric speech were examined to identify differential use of syllabic strength cues for lexical segmentation. Transcripts from listeners were divided into four groups (ranging from Better- to Poorer- performing). Phrases classified as Higher- and Lower-intelligibility were analyzed separately for each performance group to assess the independent variable of severity. Results revealed that all four listener groups used syllabic strength cues for lexical segmentation of Higher-intelligibility speech, but only the Poorer listeners persisted with this strategy for the Lower-intelligibility phrases. This finding and additional analyses suggest testable hypotheses to address the role of cue-use and performance patterns.

The effect of visual cues provided by computerised aphasia treatment

Background: In the current healthcare system that often does not allow intensive and long-term rehabilitation services, computerised aphasia treatment is considered a promising alternative. Despite the increasing interest, it is largely unknown which components of computer programs are crucial in enhancing speech-language functions of individuals with aphasia. Aims: The current study compared auditory-visual cues (i.e., listening to speech while looking at the speakers face) and auditory-only cues (i.e., listening to speech without the speakers face shown) that were presented by a computerised practice program for verbal naming. It was predicted that greater gains would be observed in the auditory-visual condition than in the auditory-only condition regardless of the severity in perceptual deficits. Methods & Procedures: Two individuals (TV & ML) with chronic aphasia and verbal apraxia practised naming 10 items with video clips (auditory-visual condition) and 10 items with sound files (auditory-only condition). The practice programs provided an increasing level of support (e.g., cues and models) to facilitate word retrieval and verbal production. Outcomes & Results: TV made more rapid and consistent improvements in the auditory-visual condition than in the auditory-only condition. His performance in the two practice conditions did not differ significantly. ML demonstrated significant improvements in both conditions. A qualitative analysis on the number of words ML spontaneously produced without a clinicians support suggested the advantage of the auditory-visual practice. Conclusions: Data from the current study suggest the advantage of including auditory-visual stimuli in computerised aphasia treatment.

A follow-up on humanoid-mediated stroke physical rehabilitation

We report the results of standardized tests on a single subject with a stroke at 4, 20 and 28 weeks after completion of the study. These results follow from previous work [1]. The subject demonstrated sustained improvement in motor function 28 weeks after completing the study. In addition to quantitative results, the questionnaire results by the subject and the spouse testify that the subjective user experience was also positive. This further advocates the use of general purpose robots to complement human therapists.0

Learning Therapy Strategies from Demonstration Using Latent Dirichlet Allocation

The use of robots in stroke rehabilitation has become a popular trend in rehabilitation robotics. However, despite the acknowledged value of customized service for individual patients, research on programming adaptive therapy for individual patients has received little attention. The goal of the current study is to model teletherapy sessions in the form of a generative process for autonomous therapy that approximate the demonstrations of the therapist. The resulting autonomous programs for therapy may imitate the strategy that the therapist might have employed and reinforce therapeutic exercises between teletherapy sessions. We propose to encode the therapists decision criteria in terms of the patients motor performance features. Specifically, in this work, we apply Latent Dirichlet Allocation on the batch data collected during teletherapy sessions between a single stroke patient and a single therapist. Using the resulting models, the therapeutic exercise targets are generated and are verified with the same therapist who generated the data.

Quantifying speech rhythm deficits in dysarthria

Disturbances in speech rhythm are common in dysarthria, and deficit patterns vary among dysarthria subtypes. Attempts to quantify differences in disordered rhythm patterns have met with limited success. Here, we attempt to differentiate various forms of dysarthria using metrics designed to quantify linguistic rhythm on a stress‐timed versus syllable‐timed continuum. Speakers with hypokinetic, ataxic, hyperkinetic, or mixed flaccid‐spastic dysarthria read sentences. Rhythm metrics were calculated for these utterances based on vocalic and consonantal interval durations. Differences among groups were discovered for %V (percentage of utterance duration comprised of vocalic intervals), VarcoV (rate‐normalized standard deviation of vocalic interval durations), and nPVI‐V (normalized pairwise variability index for vocalic intervals). Differences were independent of variable speaking rates among groups. All dysarthric groups produced rhythm patterns less stress‐timed than the control American English speakers, wi...

Adaptive therapy strategies: Efficacy and learning framework

This paper considers a data-driven framework to model target selection strategies using runtime kinematic parameters of individual patients. These models can be used to select new exercise targets that conform with the decision criteria of the therapist. We present the results from a single-subject case study with a manually written target selection function. Motivated by promising results, we propose a framework to learning customized/adaptive therapy models for individual patients. Through the data collected from a normally functioning adult, we demonstrate that it is feasible to model varying strategies from the demonstration of target selection.

Cognitive-linguistic effort in multidisciplinary stroke rehabilitation: Decreasing vs. increasing cues for word retrieval

Approximately 24% of stroke survivors experience co-occurring aphasia and hemiparesis. These individuals typically attend back-to-back therapy sessions. However, sequentially scheduled therapy may trigger physical and mental fatigue and have an adverse impact on treatment outcomes. The current study tested a hypothesis that exerting less effort during a therapy session would reduce overall fatigue and enhance functional recovery. Two stroke survivors chronically challenged by non-fluent aphasia and right hemiparesis sequentially completed verbal naming and upper-limb tasks on their home computers. The level of cognitive-linguistic effort in speech/language practice was manipulated by presenting verbal naming tasks in two conditions: Decreasing cues (i.e., most-to-least support for word retrieval), and Increasing cues (i.e., least-to-most support). The participants completed the same upper-limb exercises throughout the study periods. Both individuals showed a statistically significant advantage of decreasing cues over increasing cues in word retrieval during the practice period, but not at the end of the practice period or thereafter. The participant with moderate aphasia and hemiparesis achieved clinically meaningful gains in upper-limb functions following the decreasing cues condition, but not after the increasing cues condition. Preliminary findings from the current study suggest a positive impact of decreasing cues in the context of multidisciplinary stroke rehabilitation.