Learning Therapy Strategies from Demonstration Using Latent Dirichlet Allocation

  title={Learning Therapy Strategies from Demonstration Using Latent Dirichlet Allocation},
  author={Hee-Tae Jung and Richard Gabriel Freedman and Tammie Foster and Yu-Kyong Choe and Shlomo Zilberstein and Roderic A. Grupen},
  journal={Proceedings of the 20th International Conference on Intelligent User Interfaces},
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… 

Figures from this paper

iART: Learning From Demonstration for Assisted Robotic Therapy Using LSTM
iART presents a trajectory agnostic LfD routine that can generalize learned behavior from a single trajectory to any 3D shape, and enables the patient to modify this behavior as per their preference once the therapist's behavior has been learned.
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
Rehabilitation Games in Real-World Clinical Settings
Significant findings include different engagement patterns of stroke patients in game-assisted therapy, imperative roles of therapists in moderating games and challenges that therapists face during game- assisted therapy, and lack of support for therapists in delivering patient-centered, personalized therapy to individual stroke patients.
Rehabilitation Games in Real-World Clinical Settings: Practices, Challenges, and Opportunities
Significant findings include different engagement patterns of stroke patients in game-assisted therapy, imperative roles of therapists in moderating games and challenges that therapists face during game- assisted therapy, and lack of support for therapists in delivering patient-centered, personalized therapy to individual stroke patients.
Assisted Living for People with Disabilities using a Service Robot
This paper introduces a concept in personal robotics to support independent living for people with disabilities, and introduces the current efforts on object retrieval and physical therapies for stroke rehabilitation.
Extended Virtual Presence of Therapists through Home Service Robots
Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial
A Unifying Perspective of Plan, Activity, and Intent Recognition
A generalized formalization of recognition that can be used in all three types of recognition is proposed and a perspective that emphasizes how PAIR applies to different aspects of the generative planning and execution process is proposed.
Temporal and Object Relations in Unsupervised Plan and Activity Recognition
We consider ways to improve the performance of unsupervised plan and activity recognition techniques by considering temporal and object relations in addition to postural data. Temporal relationships
Personalized Recommendation Algorithm Based on LFM with QoS Constraint
  • Ying Jin, Yiwen Zhang
  • Computer Science
    2017 IEEE International Congress on Big Data (BigData Congress)
  • 2017
A new personalized recommendation algorithm with the quality of service (QoS) constraints based on latent factor model (LFM) is presented, capable of effectively improving the recall rate, accuracy rate and coverage rate of the personalized recommendation system.
Extended Virtual Presence of Therapists through Home Service Robots
The use of robots in rehabilitation is an increasingly viable option, given the shortage of well-trained therapists who can address individual patients' needs and priorities, but many approaches lack of adaptive therapy programs where generic exercise targets do not necessarily address the specific needs/deficit of individual patients nor impose appropriate challenges.


Adaptation of task difficulty in rehabilitation exercises based on the user's motor performance and physiological responses
  • Navid Shirzad, H. V. D. Loos
  • Psychology, Computer Science
    2013 IEEE 13th International Conference on Rehabilitation Robotics (ICORR)
  • 2013
This paper compared the performance of three machine learning algorithms in predicting a user's desirable difficulty during a typical reaching motion rehabilitation task and found that a Neural Network approach gives higher prediction accuracy in comparison with models based on k-Nearest Neighbor and Discriminant Analysis methods.
Upper extremity physical therapy for stroke patients using a general purpose robot
It is demonstrated that a general purpose robot can induce desired therapeutic exercise movements from a patient and that the challenge level can be adapted as the patient improves his motor function, which leads to observable improvements in the motor function of the patient.
Design strategies to improve patient motivation during robot-aided rehabilitation
The design of two robot devices for use in the rehabilitation of upper limb movements, that can motivate patients during the execution of the assigned motor tasks by enhancing the gaming aspects of rehabilitation are presented.
Poststroke Upper Extremity Rehabilitation: A Review of Robotic Systems and Clinical Results
A review of the current state-of-the-art in robotic applications in poststroke therapy for the upper extremity, written specifically to help clinicians determine the differences between various systems.
Effects of Robot-Assisted Therapy on Upper Limb Recovery After Stroke: A Systematic Review
Future research into the effects of robot-assisted therapy should distinguish between upper and lower robotics arm training and concentrate on kinematical analysis to differentiate between genuine upper limb motor recovery and functional recovery due to compensation strategies by proximal control of the trunk and upper limb.
Systematic review of the effect of robot-aided therapy on recovery of the hemiparetic arm after stroke.
It is found that robot-aided therapy of the proximal upper limb improves short- and long-term motor control of the paretic shoulder and elbow in subacute and chronic patients; however, it is found no consistent influence on functional abilities.
Towards ethical research practice: Anticipating social consequences of rehabilitation robots
  • Hee-Tae Jung, Danbi Yoo
  • Art
    2014 IEEE International Symposium on Ethics in Science, Technology and Engineering
  • 2014
Historically, the advent of new technology has accompanied by social issues, such as unequal access to the technology and job displacement. Rehabilitation robotics technology for post-stroke
Plan and Activity Recognition from a Topic Modeling Perspective
The application of Latent Dirichlet Allocation topic models to human skeletal data of plan execution traces obtained from a RGB-D sensor is explored and initial empirical results suggest that such NLP methods can be useful in complex PR and AR tasks.
Human Action Recognition by Semilatent Topic Models
  • Yang Wang, Greg Mori
  • Computer Science
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2009
Two new models for human action recognition from video sequences using topic models differ from previous latent topic models for visual recognition in two major aspects: first of all, the latent topics in the models directly correspond to class labels; second, some of the latent variables in previous topic models become observed in this case.
Upper-limb exercises for stroke patients through the direct engagement of an embodied agent
It is argued that a general-purpose embodied agent has a potential to functionally complement human therapists in providing rehab to stroke patients.