SlideShare a Scribd company logo
MEASURING AND QUANTIFYING QUALITY OF
MOVEMENT
Enhancing evaluation and treatment techniques for rehabilitation of hemiparetic stroke patients
Justyna Ausareny
MASTER THESIS UPF- FINAL / 2015-2016
CENTER OF AUTONOMOUS SYSTEMS AND NEURO- ROBOTICS
DEPARTMENT OF INFORMATION AND COMMUNICATION TECHNOLOGIES
1
Introduction to stroke
2
➔ Referred to as a “brain attack”.
➔ Every year first-ever stroke incidents occur about
16 million (Mathers et al. 2006)
➔ Main causes of disability in the western world
(Dobkin, 2015).
➔ 55% to 75% of stroke patients suffer from motor
dysfunctions (Larsen et al. 2005)
About 60% of post stroke patients receive inappropriate care
(Atul Gawande, 2012)
State of Art
❖ How exactly does the motor behavior change
after stroke?
3
❖ How can we can best represent it for better
evaluation and treatment?
State of Art
Introduction
to movement
4
Loss of
somatosensation
Paresis
Abnormal muscle
tone
Loss of
fractionated
movement
Movement planning
Coordination
Accuracy
Movement duration
Efficacy
Ease
Smoothness
Efficiency
Coordination
Range
State of Art
Clinical scales
Types of measures.
1) To evaluate impairment
2) Determine the impact of the impairment on daily
activities (ADL).
5
How can these impairments be addressed?
State of Art
The most popular test for motor impairment is
Fugl-Meyer Assessment (FM) (Crow et al 2008, Wei et
al 2011, Gladstone et al 2012)
Rowe/ Occupational Therapist, V. (2008, October 15). Fugl-Meyer
Assessment of Motor Recovery after Stroke [Video file]. Retrieved
from https://www.youtube.com/watch?v=B70qDfl3LyA
Clinical scales
6
➔ Robustness and Reliability
(Heppner et al. 2008 and French et al.
2001)
➔ Sensitivity
(Beebe et al 2008, Crow et al 2008,
Woodbury et al 2007, Thompson et
al 2015)
What are the issues?
How can we address
these problems?
State of Art
Problem statement
➔ clinical scales
➔ subjective tests
➔ difficult to generalize
7
So far, the Rehabilitation Gaming System Wearable (RGS-wear) has tackled
this issue by providing a wearable-based solution as an intervention tool,
focusing on the amount of movement.
State of Art
In this thesis, the aim is to enhance the current state and establish a novel
measurement providing qualitative assessment of movement
➔ Ability to objectively and repeatedly measure movement in a way that
quantify motor recovery
Technology
8
State of Art
➔ Low cost
➔ Provide more access for measuring movement in an evaluation activity
➔ Accelerometer has been applied for the evaluation of stroke motor
recovery (Akay et al. 2003; Bonato et al. 2004).
➔ Allows for motor pattern analysis (Giansanti 2003).
Motion evaluation
9
State of Art
Motion analysis systems are collections of
algorithms that transform the data from
one abstraction level to another
(Sant'Anna and Wickstrom 2013).
These systems address 2 main questions:
1. What is being performed?
2. How the activity is being performed?
Retrieved from
http://what-when-how.com/medical-informatics/i
Assessment and treatment
➔ Reaching task is one of the
most popular activities has
been extensively studied also
present in everyday activities
ADL (Wagner et al 2007).
➔ Circle drawing is a relatively
new method that has shown to
be suitable for making
distinction between healthy
subjects and stroke survivors.
(Krabben et al 2011)
10
State of Art
➔ Task-oriented training can enhance functional
plasticity in cortical organization (Chollet et al.
1991).
➔ The key elements in motor recovery are quantity
and task specificity (Cirstea et al. 2007).
Retrieved from http://rgs-project.eu/sites/all/themes/danland/images/slideshows/header1.jpg
Hypothesis
The purpose of this study is to
11
Methods
1. Provide a technical validation of a methodology for the analysis of quality of movement of the hand,
based on data from inertial motion units (accelerometer and gyroscope), in 2 specific tasks: planar
reaching and planar circle drawing.
We hypothesize that our system/methodology for the analysis of quality of
movement is able to capture differences in performance and identify
movement deficits associated with stroke.
Apparatus
Primary system hardware:
➔ Open Metawear wearable (model RG)
with 6 axis of motion sensing
12
Methods
Secondary system hardware for
validation of position of Metawear:
➔ Reactable
Subjects
24 healthy subjects recruited from UPF
Poblenou campus. All subjects signed a
consent and approved the data acquisition
process.
Methods
Experiment
14
Methods
➔ Task 1. Reaching
➔ Task 2. Circle Drawing
Procedure
15
Methods
Reactable
16
Methods
Methods linear acceleration
17
Methods
Gravity compensation using AHRS sensor fusion algorithm (Seb Madgwick, 2013)
Retrieved from
http://what-when-how.com/medical-informatics/i
Methodstrajectories and trails
18
Methods
Trajectory reconstruction using modified AHRS sensor
fusion algorithm (Seb Madgwick, 2013)
Trail detection
Trial detection
Methods preliminary analysis
19
Methods
Example of a detected trial and analysis
Measures
20
Methods
Loss of
somatosensation
Paresis Abnormal muscle
tone
Loss of fractionated
movement
References:
Nordin et al (2011)
Iwamuro et al (2008)
Chang et al (2007)
Johnson (2011)
Balasubramania, (2015)
Results Dominant vs Non Dominant
21
Results
Results Non Constrained vs Constrained
22
Results
Correlations
23
Correlations
24
Correlations
25
Correlations
26
27
Non constrained Constrained
Reaching Circle drawing Reaching Circle drawing
Dominant Non dominant Dominant Non dominant Dominant Non dominant Dominant Non dominant
Ease -
Temporal
Efficiency
0.0018 0.0001 0.0947 8.1947 e -0.5 0.0023 0.1052 0.127 0.0310
Smoothness -
Ease
0.0037 0.0134 2.5341 e -0.5 1.561 e -0.5 5.3489 e -0.5 1.3442 e -0.5 1.9739 e -0.5 8.8993 e -0.9
Smoothness -
Movement
Planning
4.9044 e -0.5 0.0134 1.0138 e -0.9 5.1486 e -0.6 0.0006 0.0155 0.0002 3.6502 e -0.7
Smoothness -
Temporal
efficiency
0.0037 0.0134 2.5341 e -0.5 1.561 e -0.5 5.3489 e -0.5 1.3442 e -0.5 1.9739 e -0.5 8.8993 e -0.9
Movement
Planning -
Temporal
efficiency
1.7202 e -0.5 4.1917 e -0.9 1.074 e -0.6 4.5677 e -0.9 0.0004 3.2163 e -0.7 2.4044 e -0.8 0.0002
Ease -
Movement
Planning
0.0650 0.0004 0.0572 0.003 0.0157 0.1047 0.1271 0.1313
P
values
Conclusions
28
➔ We can objectively assess the quality of movement and repeatedly measure
it
➔ When evaluating only healthy subjects the difficulty of task must be taken
into account
➔ The motion systems should sample data at high frequency in order to
insure accuracy
➔ The connectivity and ease of use are the current constraints in the study
that should be addressed in future directions
The end
Thank you !
29
Usability
30
➔ Overall Satisfaction and Ergonomics
➔ User Experiment - Interface Satisfaction
➔ Wearable System Usability Scale
➔ Perceived Usefulness of the Study
➔ Social and Technical Human Factors
Suggestions for future developments:
➔ Add heart rate skin conductance
➔ Different study populations
➔ Implementable devices
➔ Connectivity
➔ Easier to wear
➔ Instant feedback
Overall Satisfaction and Ergonomics
31
Agree
Agree
User Experiment - Interface Satisfaction
32
Comfortable
Wearable System Usability Scale
33ComfortableAgreeAgree
AgreeAgree
Perceived Usefulness & Socio -Technical Human Factors
34
Agree
Was there anything about the tool that surprised you?
Over 50% responded no,
Easier to wear
Not challenging
Discomfort when using only 1 hand
Smartphone
Uncomfortable bands
How did it feel for you to have your movement assessed by a
computer?
Over 50% responded fine,
More careful
Uncomfortable “felt like a robot”
Missing feedback
How did it feel to have no physical cues (body language
or verbal communication)
Over 50% responded fine,
Aroused curiosity
Similar to online course
Eliminates pressure of social interaction
Better for concentration
Boring “could have done a lot more things face to face”

More Related Content

PDF
The Effectiveness of Mirror Therapy with Stroke Patients in Producing Improve...
PDF
ANALYSING THE CORRELATION OF GERIATRIC ASSESSMENT SCORES AND ACTIVITY IN SMAR...
PDF
Ergonomics Analysis of Blanket Lifting Technique Using Posture Evaluation In...
PDF
Estimation of Walking rate in Complex activity recognition
PDF
Increase nursing time spent with patients: the holy grail. A mixed method res...
PPTX
Rehabilitation and Outcomes - Year in Review
PDF
ANALYSING THE CORRELATION OF GERIATRIC ASSESSMENT SCORES AND ACTIVITY IN SMAR...
PPTX
Wallaroo with Isonas Presentation
The Effectiveness of Mirror Therapy with Stroke Patients in Producing Improve...
ANALYSING THE CORRELATION OF GERIATRIC ASSESSMENT SCORES AND ACTIVITY IN SMAR...
Ergonomics Analysis of Blanket Lifting Technique Using Posture Evaluation In...
Estimation of Walking rate in Complex activity recognition
Increase nursing time spent with patients: the holy grail. A mixed method res...
Rehabilitation and Outcomes - Year in Review
ANALYSING THE CORRELATION OF GERIATRIC ASSESSMENT SCORES AND ACTIVITY IN SMAR...
Wallaroo with Isonas Presentation

What's hot (19)

PPTX
Salon 1 14 kasim 15.30 17.00 duygu demi̇r-ing
PDF
Improving the performance of k nearest neighbor algorithm for the classificat...
PPT
Hemiparesis
DOCX
The Effect of Imagery on 1RM Scores of College Students
PDF
IRJET- Diabetes Prediction by Machine Learning over Big Data from Healthc...
PPTX
Activ pal technologies
PPTX
Exergames for Patients in Acute Care Settings: Systematic Review of the Repor...
PDF
FORESTALLING GROWTH RATE IN TYPE II DIABETIC PATIENTS USING DATA MINING AND A...
PPT
National Strategy for Improving Outcomes for Sudden Cardiac Arrest in Singapore
PPT
Clinical Prediction Rules
PDF
Effective of a Structured Teaching Module Regarding Care of Children in the C...
PPTX
PDF
Let's Talk Research Annual Conference - 24th-25th September 2014 (Paula Bennett)
PDF
Brigid Gillespie - Griffith University
PDF
Health IT for Executives (March 28, 2017)
PDF
PERFORMANCE OF DATA MINING TECHNIQUES TO PREDICT IN HEALTHCARE CASE STUDY: CH...
PPT
Royal Inland Hospital Emergency OR usage (February 2012)
PDF
Exercícios controle motor e lombalgia
PDF
Francesca Martinelli MedicReS World Congress 2013
Salon 1 14 kasim 15.30 17.00 duygu demi̇r-ing
Improving the performance of k nearest neighbor algorithm for the classificat...
Hemiparesis
The Effect of Imagery on 1RM Scores of College Students
IRJET- Diabetes Prediction by Machine Learning over Big Data from Healthc...
Activ pal technologies
Exergames for Patients in Acute Care Settings: Systematic Review of the Repor...
FORESTALLING GROWTH RATE IN TYPE II DIABETIC PATIENTS USING DATA MINING AND A...
National Strategy for Improving Outcomes for Sudden Cardiac Arrest in Singapore
Clinical Prediction Rules
Effective of a Structured Teaching Module Regarding Care of Children in the C...
Let's Talk Research Annual Conference - 24th-25th September 2014 (Paula Bennett)
Brigid Gillespie - Griffith University
Health IT for Executives (March 28, 2017)
PERFORMANCE OF DATA MINING TECHNIQUES TO PREDICT IN HEALTHCARE CASE STUDY: CH...
Royal Inland Hospital Emergency OR usage (February 2012)
Exercícios controle motor e lombalgia
Francesca Martinelli MedicReS World Congress 2013
Ad

Similar to Measuring and quantifying quality of movement (20)

PDF
Important Parameters for Hand Function Assessment of Stroke Patients
PPTX
Tcs tele rehab-hod-0.4
PPTX
A Hierarchical Approach to Recognize Purposeful Movements Using Inertial Sens...
PPT
Advancing-Mobility-Through-Progressive-Technology.ppt
PPTX
IP Rehab: The Latest Evidence in rehab.pptx
PDF
A systematic approach towards designing low-cost motor and cognitive rehabili...
PDF
Step by-step project - demo
PDF
Sistema di misura per la valutazione del cammino
PPTX
Stroke
PPTX
Motor control Assessment
PPTX
Technology currently using in Rehab.pptx
PDF
Active model
PPT
Pre-Operative and Post-Operative Assessments.ppt
PDF
Abstract_Giovanni_Mascia_196865
PPT
robotic and rehabilitation in medicine.ppt
PPTX
Wii Fit balance training in outpatient people with subacute stroke
PPTX
Use of technology in rehabilitation - Lorna Paul
PPTX
Functional assessment scale.pptx
PPTX
1. Introduction_BPT_Neurophysio_Preeti.pptx
PPTX
1. Introduction_BPT_Neurophysio_Preeti.pptx
Important Parameters for Hand Function Assessment of Stroke Patients
Tcs tele rehab-hod-0.4
A Hierarchical Approach to Recognize Purposeful Movements Using Inertial Sens...
Advancing-Mobility-Through-Progressive-Technology.ppt
IP Rehab: The Latest Evidence in rehab.pptx
A systematic approach towards designing low-cost motor and cognitive rehabili...
Step by-step project - demo
Sistema di misura per la valutazione del cammino
Stroke
Motor control Assessment
Technology currently using in Rehab.pptx
Active model
Pre-Operative and Post-Operative Assessments.ppt
Abstract_Giovanni_Mascia_196865
robotic and rehabilitation in medicine.ppt
Wii Fit balance training in outpatient people with subacute stroke
Use of technology in rehabilitation - Lorna Paul
Functional assessment scale.pptx
1. Introduction_BPT_Neurophysio_Preeti.pptx
1. Introduction_BPT_Neurophysio_Preeti.pptx
Ad

Recently uploaded (20)

PDF
Univ-Connecticut-ChatGPT-Presentaion.pdf
PDF
Transform Your ITIL® 4 & ITSM Strategy with AI in 2025.pdf
PPTX
Group 1 Presentation -Planning and Decision Making .pptx
PDF
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
PDF
DASA ADMISSION 2024_FirstRound_FirstRank_LastRank.pdf
PDF
Developing a website for English-speaking practice to English as a foreign la...
PDF
Getting started with AI Agents and Multi-Agent Systems
PDF
A review of recent deep learning applications in wood surface defect identifi...
PDF
Hindi spoken digit analysis for native and non-native speakers
PPTX
The various Industrial Revolutions .pptx
PDF
August Patch Tuesday
PDF
Enhancing emotion recognition model for a student engagement use case through...
PDF
A novel scalable deep ensemble learning framework for big data classification...
PDF
DP Operators-handbook-extract for the Mautical Institute
PDF
Hybrid model detection and classification of lung cancer
PDF
NewMind AI Weekly Chronicles – August ’25 Week III
PDF
A Late Bloomer's Guide to GenAI: Ethics, Bias, and Effective Prompting - Boha...
PDF
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
PDF
A contest of sentiment analysis: k-nearest neighbor versus neural network
PDF
From MVP to Full-Scale Product A Startup’s Software Journey.pdf
Univ-Connecticut-ChatGPT-Presentaion.pdf
Transform Your ITIL® 4 & ITSM Strategy with AI in 2025.pdf
Group 1 Presentation -Planning and Decision Making .pptx
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
DASA ADMISSION 2024_FirstRound_FirstRank_LastRank.pdf
Developing a website for English-speaking practice to English as a foreign la...
Getting started with AI Agents and Multi-Agent Systems
A review of recent deep learning applications in wood surface defect identifi...
Hindi spoken digit analysis for native and non-native speakers
The various Industrial Revolutions .pptx
August Patch Tuesday
Enhancing emotion recognition model for a student engagement use case through...
A novel scalable deep ensemble learning framework for big data classification...
DP Operators-handbook-extract for the Mautical Institute
Hybrid model detection and classification of lung cancer
NewMind AI Weekly Chronicles – August ’25 Week III
A Late Bloomer's Guide to GenAI: Ethics, Bias, and Effective Prompting - Boha...
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
A contest of sentiment analysis: k-nearest neighbor versus neural network
From MVP to Full-Scale Product A Startup’s Software Journey.pdf

Measuring and quantifying quality of movement

  • 1. MEASURING AND QUANTIFYING QUALITY OF MOVEMENT Enhancing evaluation and treatment techniques for rehabilitation of hemiparetic stroke patients Justyna Ausareny MASTER THESIS UPF- FINAL / 2015-2016 CENTER OF AUTONOMOUS SYSTEMS AND NEURO- ROBOTICS DEPARTMENT OF INFORMATION AND COMMUNICATION TECHNOLOGIES 1
  • 2. Introduction to stroke 2 ➔ Referred to as a “brain attack”. ➔ Every year first-ever stroke incidents occur about 16 million (Mathers et al. 2006) ➔ Main causes of disability in the western world (Dobkin, 2015). ➔ 55% to 75% of stroke patients suffer from motor dysfunctions (Larsen et al. 2005) About 60% of post stroke patients receive inappropriate care (Atul Gawande, 2012) State of Art
  • 3. ❖ How exactly does the motor behavior change after stroke? 3 ❖ How can we can best represent it for better evaluation and treatment? State of Art
  • 4. Introduction to movement 4 Loss of somatosensation Paresis Abnormal muscle tone Loss of fractionated movement Movement planning Coordination Accuracy Movement duration Efficacy Ease Smoothness Efficiency Coordination Range State of Art
  • 5. Clinical scales Types of measures. 1) To evaluate impairment 2) Determine the impact of the impairment on daily activities (ADL). 5 How can these impairments be addressed? State of Art The most popular test for motor impairment is Fugl-Meyer Assessment (FM) (Crow et al 2008, Wei et al 2011, Gladstone et al 2012) Rowe/ Occupational Therapist, V. (2008, October 15). Fugl-Meyer Assessment of Motor Recovery after Stroke [Video file]. Retrieved from https://www.youtube.com/watch?v=B70qDfl3LyA
  • 6. Clinical scales 6 ➔ Robustness and Reliability (Heppner et al. 2008 and French et al. 2001) ➔ Sensitivity (Beebe et al 2008, Crow et al 2008, Woodbury et al 2007, Thompson et al 2015) What are the issues? How can we address these problems? State of Art
  • 7. Problem statement ➔ clinical scales ➔ subjective tests ➔ difficult to generalize 7 So far, the Rehabilitation Gaming System Wearable (RGS-wear) has tackled this issue by providing a wearable-based solution as an intervention tool, focusing on the amount of movement. State of Art In this thesis, the aim is to enhance the current state and establish a novel measurement providing qualitative assessment of movement
  • 8. ➔ Ability to objectively and repeatedly measure movement in a way that quantify motor recovery Technology 8 State of Art ➔ Low cost ➔ Provide more access for measuring movement in an evaluation activity ➔ Accelerometer has been applied for the evaluation of stroke motor recovery (Akay et al. 2003; Bonato et al. 2004). ➔ Allows for motor pattern analysis (Giansanti 2003).
  • 9. Motion evaluation 9 State of Art Motion analysis systems are collections of algorithms that transform the data from one abstraction level to another (Sant'Anna and Wickstrom 2013). These systems address 2 main questions: 1. What is being performed? 2. How the activity is being performed? Retrieved from http://what-when-how.com/medical-informatics/i
  • 10. Assessment and treatment ➔ Reaching task is one of the most popular activities has been extensively studied also present in everyday activities ADL (Wagner et al 2007). ➔ Circle drawing is a relatively new method that has shown to be suitable for making distinction between healthy subjects and stroke survivors. (Krabben et al 2011) 10 State of Art ➔ Task-oriented training can enhance functional plasticity in cortical organization (Chollet et al. 1991). ➔ The key elements in motor recovery are quantity and task specificity (Cirstea et al. 2007). Retrieved from http://rgs-project.eu/sites/all/themes/danland/images/slideshows/header1.jpg
  • 11. Hypothesis The purpose of this study is to 11 Methods 1. Provide a technical validation of a methodology for the analysis of quality of movement of the hand, based on data from inertial motion units (accelerometer and gyroscope), in 2 specific tasks: planar reaching and planar circle drawing. We hypothesize that our system/methodology for the analysis of quality of movement is able to capture differences in performance and identify movement deficits associated with stroke.
  • 12. Apparatus Primary system hardware: ➔ Open Metawear wearable (model RG) with 6 axis of motion sensing 12 Methods Secondary system hardware for validation of position of Metawear: ➔ Reactable
  • 13. Subjects 24 healthy subjects recruited from UPF Poblenou campus. All subjects signed a consent and approved the data acquisition process. Methods
  • 14. Experiment 14 Methods ➔ Task 1. Reaching ➔ Task 2. Circle Drawing
  • 17. Methods linear acceleration 17 Methods Gravity compensation using AHRS sensor fusion algorithm (Seb Madgwick, 2013) Retrieved from http://what-when-how.com/medical-informatics/i
  • 18. Methodstrajectories and trails 18 Methods Trajectory reconstruction using modified AHRS sensor fusion algorithm (Seb Madgwick, 2013) Trail detection Trial detection
  • 19. Methods preliminary analysis 19 Methods Example of a detected trial and analysis
  • 20. Measures 20 Methods Loss of somatosensation Paresis Abnormal muscle tone Loss of fractionated movement References: Nordin et al (2011) Iwamuro et al (2008) Chang et al (2007) Johnson (2011) Balasubramania, (2015)
  • 21. Results Dominant vs Non Dominant 21 Results
  • 22. Results Non Constrained vs Constrained 22 Results
  • 27. 27 Non constrained Constrained Reaching Circle drawing Reaching Circle drawing Dominant Non dominant Dominant Non dominant Dominant Non dominant Dominant Non dominant Ease - Temporal Efficiency 0.0018 0.0001 0.0947 8.1947 e -0.5 0.0023 0.1052 0.127 0.0310 Smoothness - Ease 0.0037 0.0134 2.5341 e -0.5 1.561 e -0.5 5.3489 e -0.5 1.3442 e -0.5 1.9739 e -0.5 8.8993 e -0.9 Smoothness - Movement Planning 4.9044 e -0.5 0.0134 1.0138 e -0.9 5.1486 e -0.6 0.0006 0.0155 0.0002 3.6502 e -0.7 Smoothness - Temporal efficiency 0.0037 0.0134 2.5341 e -0.5 1.561 e -0.5 5.3489 e -0.5 1.3442 e -0.5 1.9739 e -0.5 8.8993 e -0.9 Movement Planning - Temporal efficiency 1.7202 e -0.5 4.1917 e -0.9 1.074 e -0.6 4.5677 e -0.9 0.0004 3.2163 e -0.7 2.4044 e -0.8 0.0002 Ease - Movement Planning 0.0650 0.0004 0.0572 0.003 0.0157 0.1047 0.1271 0.1313 P values
  • 28. Conclusions 28 ➔ We can objectively assess the quality of movement and repeatedly measure it ➔ When evaluating only healthy subjects the difficulty of task must be taken into account ➔ The motion systems should sample data at high frequency in order to insure accuracy ➔ The connectivity and ease of use are the current constraints in the study that should be addressed in future directions
  • 30. Usability 30 ➔ Overall Satisfaction and Ergonomics ➔ User Experiment - Interface Satisfaction ➔ Wearable System Usability Scale ➔ Perceived Usefulness of the Study ➔ Social and Technical Human Factors Suggestions for future developments: ➔ Add heart rate skin conductance ➔ Different study populations ➔ Implementable devices ➔ Connectivity ➔ Easier to wear ➔ Instant feedback
  • 31. Overall Satisfaction and Ergonomics 31 Agree Agree
  • 32. User Experiment - Interface Satisfaction 32 Comfortable
  • 33. Wearable System Usability Scale 33ComfortableAgreeAgree AgreeAgree
  • 34. Perceived Usefulness & Socio -Technical Human Factors 34 Agree Was there anything about the tool that surprised you? Over 50% responded no, Easier to wear Not challenging Discomfort when using only 1 hand Smartphone Uncomfortable bands How did it feel for you to have your movement assessed by a computer? Over 50% responded fine, More careful Uncomfortable “felt like a robot” Missing feedback How did it feel to have no physical cues (body language or verbal communication) Over 50% responded fine, Aroused curiosity Similar to online course Eliminates pressure of social interaction Better for concentration Boring “could have done a lot more things face to face”