Top Banner
Making sense of movement with the virtual Prof: development of a cross-faculty set of online tutorials Background QMUL students from a range of disciplines struggle to make the most of complex movement capture data sets, limiting outcomes of their teaching and research. Furthermore, we are missing the chance to develop key transferable analysis skills. We therefore proposed a cross-faculty collaboration to develop a suite of e-learning resources, to act as a 'virtual prof' and yield a long-term solution. Funding was requested to develop these resources in an iterative process and to share them. Aims The aim of the project was to solve the problem faced by many students in analysing their complex movement capture data. Our objectives were to develop a step-by-step e-learning resource through QM+ that would initially be comprised of a composition of short tutorials, in addition to a collation of existing on-line resources with the aim of enabling the student to work with a basic text file to transform it to a graphical visualisation that includes kinetic analysis and synchronisation of muscle activation (electromyography) with kinematic and kinetic moment outputs from movement data. It was envisaged that the impact of this work would be significant transferable skills, improved data analysis, improvement of resources to support teaching and research and a set of transferable electronic resources which will contribute to QMUL's support of students in handling complex data sets. These have all been realised. Activities Two students were supported by Professor Woledge and others in SEM to draft initial learning resources. These drafts were then refined, and added to by Dr Sarah Domone and Dr Birn-Jeffrey to deliver the core material. Dr Saira Chaudhry then adapted these and uploaded them on to QM+. A well-attended engagement and dissemination event was held as part of a week of events focussing on muscles and movement, open to a wide audience of people interested in motion capture. This hackathon introduced motion capture experts with motion capture students to solve their problems and was well received. Three events that week were held, and structured as tribute to Professor Roger Woledge who died suddenly in a horse riding accident. Outcomes The original aims of the project were (Figure 1) in the main, met, although we inevitably struggled to deliver these to time.
17

Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Jul 22, 2020

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Making sense of movement with the virtual Prof: development of

a cross-faculty set of online tutorials Background

QMUL students from a range of disciplines struggle to make the most of complex movement capture

data sets, limiting outcomes of their teaching and research. Furthermore, we are missing the chance

to develop key transferable analysis skills. We therefore proposed a cross-faculty collaboration to

develop a suite of e-learning resources, to act as a 'virtual prof' and yield a long-term solution.

Funding was requested to develop these resources in an iterative process and to share them.

Aims

The aim of the project was to solve the problem faced by many students in analysing their complex

movement capture data. Our objectives were to develop a step-by-step e-learning resource through

QM+ that would initially be comprised of a composition of short tutorials, in addition to a collation

of existing on-line resources with the aim of enabling the student to work with a basic text file to

transform it to a graphical visualisation that includes kinetic analysis and synchronisation of muscle

activation (electromyography) with kinematic and kinetic moment outputs from movement data. It

was envisaged that the impact of this work would be significant transferable skills, improved data

analysis, improvement of resources to support teaching and research and a set of transferable

electronic resources which will contribute to QMUL's support of students in handling complex data

sets. These have all been realised.

Activities

Two students were supported by Professor Woledge and others in SEM to draft initial learning

resources. These drafts were then refined, and added to by Dr Sarah Domone and Dr Birn-Jeffrey to

deliver the core material. Dr Saira Chaudhry then adapted these and uploaded them on to QM+. A

well-attended engagement and dissemination event was held as part of a week of events focussing

on muscles and movement, open to a wide audience of people interested in motion capture. This

hackathon introduced motion capture experts with motion capture students to solve their problems

and was well received. Three events that week were held, and structured as tribute to Professor

Roger Woledge who died suddenly in a horse riding accident.

Outcomes

The original aims of the project were (Figure 1) in the main, met, although we inevitably struggled to

deliver these to time.

Page 2: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Figure 1: original plan

Specifically, we managed to:

1. Capture student learning with regards to analysis of data;

2. Develop e-learning system to teach analysis of complex data sets, i.e. from ‘movement to meaning;

3. Maximize student input through their participation in development of the teaching resource;

4. Build skill development in handling large data sets – key skill for their future careers;

5. Train students in tackling unfamiliar problems;

6. Provide a stimulating method for skills development in data analysis

We also managed to run the dissemination events. Our next step is to make sure the resources are

well used by students this year.

Challenges

There were two particular challenges. The first was the impact of Roger’s death which necessitated

us to take more time and find another data analyst to assist. The second barrier has been getting the

resources in a suitable form, now achieved via QM+ as planned, which has limited our ability to get

feedback – planned for this academic year.

Page 3: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Evaluation

Students were assessed through survey and focus groups in order to develop the resource. We had

some formal feedback from the hackathon (appendix 1), although the response rate to this was

limited. Using these resources this year will garner more feedback.

Students from different disciplinary background will be informed of the Virtual Prof resource.

Although open to all, we will target MEng students who will have some basic knowledge in

programming and/or specifically using MatLab. Students from SEM will be unlikely to have had any

formal previous programming training. Therefore, students from these different disciplines will give

us a more comprehensive feedback across both new and more advanced programming users.

Focus groups consisting of a) 3 students who used the resource extensively, and b) 3 students who

used the resource lightly but could have benefitted from using it more, will be formed each year. The

focus groups will be asked about the usefulness of the package, whether it helped them, what the

strengths/weaknesses were and whether they have any suggestions for improving the tool. From

this input, the resource will be continually updated on at least a yearly basis to improve the

effectiveness of the tool.

Enrolling MEng students to give feedback, as they have some basic knowledge regarding Matlab and

programming

SEM student will be directed to use this as a resource during their data analysis phase and feedback

will be collected from them on what they found useful, how it helped them, what are the strength

and weakness of Virtual Prof and any suggestions for improvement.

SEM students will be informed of this package on the Research methods module during their data

analysis phase. A link to the QM+ page will also be made available through the Research methods

page. Project supervisors will be informed of this tool, and asked to direct students to this resource.

Further work and dissemination

Currently not all project supervisors are aware of the Virtual Prof tool on the QM+ website.

Therefore to disseminate the access to this tool, an email flyer will be generated (similar to the flyer

created for the “Hackathon event”), and sent round to all supervisors within the relevant groups.

This flyer will also be used to advertise the tool more broadly across Queen Mary. It is hoped that all

students who require biomechanical analysis, will be directed to the QM+ resource in their data

analysis phase.

The project will be presented in the Educational Research Seminar and will be shared on the

teaching and learning website under ‘’ Case Studies and Developments’’ section that contains

updates and commentaries on projects and innovations carried out at Queen Mary. In addition to

this, the tutorials will also be launched online on the QMPlus website for students to access across

the college (http://qmplus.qmul.ac.uk/course/view.php?id=5161).

Page 4: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Appendix 1: Hackathon planning details

The Roger Woledge

Hackathon Organised by Queen Mary University

Making sense of movement with the Virtual Prof Website address

First Multidisciplinary Hackathon at the

QMUL East London

Speed dating, computing support,

analyst stations, statistical solution,

networking, industry involvement, new

technologies and other activities

Focus on muscles and movement and

win prizes

23rd April 2016

at

Queen Mary

University of London

Page 5: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear
Page 6: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Meeting challenges and developing solutions across boundaries. Saturday April 23rd 2016, QMUL East London

Editorial We will connect researchers, experts, students, analysts, scientists, inventors, business

leaders and the coalface crew in an innovative format designed to share expertise and solve problems

– on the day, before the day and beyond.

We will provide space and support to solve problems.

Professor Emeritus Roger Woledge was the quintessential scientist, academic, mentor and

educator. He applied his educational ethos and analytical expertise to a diversity of problems, and to

the benefit of varied research, translational and educational groups. He also made amazing

marmalade, gave wise counsel and modelled how to treat people. His legacy includes generations of

clinicians, scientists and industrial innovators who get better clinical results, answer important

questions and produce amazing inventions. All the while valuing each other as complete human

beings. What’s not to inspire?

The day will involve what?

Speed dating, Computing support, Analyst stations, Guidance by experts Statistical solutions, Networking lunch, Industry involvement, Animations…

Possible Organisers:

Dylan and TEAM – human biomech

Tim and team – veterinary bomech

Pat H – media and arts

Akram A – sensors

Sean Williams – statistics

Industry lead – from EECS

David Arrowsmith - Maths

Page 7: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Synopsis

Dates: 23rd April 2016 from 9am to 6pm. And it continues after at the pub…

Location: Room G27 of the School of Engineering and Materials Science

(SEMS) – Bancroft Road, London.

Access: Stepney green station for London underground

Queen Mary University bus stop

Price: £20

Synopsis of the event

TIME ACTIVITY

09:00

09:30

Registration - breakfast

10:00

10:30

Opening – Chair speech and introduction of the experts

10:30

11:00 Announce problems and speed dating

11:00

11:30

Match Groups –Team building- and start working

11:30

13:00

Working in team – Group 1–

13:00 13:30

Lunch Break

13:30

15:00

Working in team – Group 2–

15:00

Working in team – Group 3–

Page 8: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

16:30

16:30

17:00

Last hacks, finish up your work

17:00 18:00

Presentation/ demo / prizes

18:00 late

Post event Pub

Email address of organisers: [email protected] [email protected]

[email protected]

Page 9: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Administrative information:

Event title The Roger Woledge Hackathon

Event subtitle Making sense of movement with virtual prof

Event date 23rd April 2016

Event timings – START AND END

9.00 to 21.00

Event venue Room G27 of the School of Engineering and Materials Science (SEMS) – Bancroft Road – Stepney green underground station

Synopsis of event See below

Registration information

Surname Firstname Email address Apply as – a company/ a university/ a researcher/ a student/ a technician/ other Organisation: Country Address City Zip code Website: Profil: Maths/Engineer/Developer/Health Professional/ Entrepreneur Any food diet :

Preliminary flyer

Number of places

Approximatively 60 attendees

URL of the web page

In process

Special requirements

Laptops for each table in the G27 room ?

Appendix 2:

Page 10: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Hackathon summary

Presentation of the Virtual Prof The Virtual Prof is a series of tutorials to learn how to program under Matlab (MathWorks) for

students. The 13 chapters allows us to learn the basics of the signal processing tool box, from the

Matlab layout to several statistics tool such as the linear regression. Thus the student will

understand the syntax of this programming language and the logics behind the analysis. Thanks to

examples, it is possible to them to copy+paste few lines of code and to change them for seeing what

happens then. The 13 chapters are:

1. Layout of the Workspace

2. Reading and Writing files

3. Cleaning, aligning and filtering data

4. Vector and matrices

5. Storing and Naming data

6. Graphs

7. Normalization

8. 3D orientation and rotation matrices

9. 3D transformations

10. Distributions and transformation of data set

11. Measures of central tendency

12. Correlation and linear regression

13. Basic comparative stats

A page on QM+ has been set up with each tutorial material as embedded and available online

(figures below)

Page 11: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear
Page 12: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear
Page 13: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear
Page 14: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear
Page 15: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear
Page 16: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear
Page 17: Making sense of movement with the virtual Prof ... · 9. 3D transformations 10. Distributions and transformation of data set 11. Measures of central tendency 12. Correlation and linear

Report on the Hackathon Queen Mary University of London hosted the first Roger Woledge Hackathon on the 23rd of April 2016.

Researchers, experts, PhD students and business leaders were present for about 25 attendees. It was

an opportunity to solve complex problems in a multidisciplinary environment. Following the main

theme ‘Making sense of movement with the virtual prof’, attendees were encouraged to work as

teams around a shared problem and develop algorithms. Works related to EMG data, accelerometers,

gait analysis and imaging were presented under the supervision of Dr Dylan Morrissey and the invited

experts:

1. Dr Glenn Lichtwark – Biomechanics and Muscle Physiology – University of Queensland

2. Dr Monica Daley – Locomotor Biomechanics – Royal Veterinary College

3. Dr Jim Usherwood – terrestrial and aerial locomotion – Royal Veterinary College

4. Dr Emma Hodson-Tole – Neuromusculoskeletal Integration – Manchester Metropolitan

University

5. Dr Bertrand Bru – accelerometers – Charnwood Dynamics Ltd

In a friendly atmosphere and around pizzas, the participants were able to discuss about their project

and share their idea before to make them concrete. The good feedback we received encourage us to

reiterate the experience next year. For the next edition, we will take into account the highlights which

will be provided by the survey sent to the attendees.

Finally, the day finished in the pub to enable conversations to continue. Watch out for the next event.