LOGO-SelfBack

Self-management of low back pain

SELFBACK is an EU funded Horizon 2020 project to develop a monitoring system to assist patients  to self-mange low back pain. The system uses wearable sensors to continuously monitors users’ physical activity and sleep, and mobile phones to provide decision support and behavioural change interventions to the user. The system aims  to recommend personalised self-management plans to users while automatically tracking adherence of users to recommended plans.
RGU’s Contribution
RGU’s contribution to the project is the development of activity recognition algorithms and the development of theory-backed digital intervention approaches.


selfback_home

nw

Professor Nirmalie Wiratunga
Professor Wiratunga’s research interests include both theoretical and practical aspects of Artificial Intelligence with focus on Case-based Reasoning (CBR), Text Mining and Machine Learning. Her recent projects include: an EU H2020 Selfback project on mHealth and wearables; InnovateUK projects on Sentiment Analysis for Product recommendation, Diabetes Self-Management, Knowledge Discovery from ePatient Records and industry funded projects with BT and British Geological Survey.
sm

Dr Stewart Massie

Dr Massie has more than 10 years research experience in Artificial Intelligence developing improved machine learning, information retrieval and data mining technologies with a focus on the application of introspective and other learning techniques. His main expertise includes case-based reasoning and recommendation, with recent funded projects developing health and tourism applications.
kc

Dr Kay Cooper
Dr Cooper is a Reader in Health & Wellbeing Research with over 10 years’ experience of research in the fields of musculoskeletal conditions and physiotherapy service delivery. She leads the North of Scotland hub of the Council for Allied Health Professions Research, has served as an expert panel member for the Chartered Society of Physiotherapy’s research priorities exercise, and is a regular reviewer for several peer-reviewed journals and conferences in the field of physiotherapy.
ss

Dr Sadiq Sani
Dr Sani is a Research Fellow who specialises in machine learning, case-based reasoning and natural language processing. He has experience working on diverse projects, from search personalisation to computer vision and more recently, human activity recognition. Sadiq presently works on the SelfBACK project, developing activity recognition algorithms to assist with self-management of low back pain.

2017

Learning Deep and Shallow Features for Human Activity Recognition
Sani, S., Massie, S., Wiratunga, N. and Cooper, K.
International Conference on Knowledge Science, Engineering and Management – KSEM 2017
knn Sampling for Personalised Human Activity Recognition
Sani, S., Wiratunga, N., Massie, S. and Cooper, K.
International Conference on Case-Based Reasoning – ICCBR 2017
Learning Deep Features for kNN-Based Human Activity Recognition
Sani, S., Wiratunga, N., Massie, S.
Workshop on Case-based Reasoning and Deep Learning – CBRDL 2017
A Siamese Convolutional Network for Developing Similarity Knowledge in the SelfBACK Dataset
Martin, K., Wiratunga, N., Sani, S., Massie, S. and Clos, J.
Workshop on Case-based Reasoning and Deep Learning – CBRDL 2017
Accuracy of Physical Activity Recognition from a Wrist-worn Sensor
Cooper, K., Sani, S., Corrigan, L., MacDonald, H., Prentice, C., Vareta, R., Massie, S., Wiratunga, N.
Physiotherapy UK 2017

2016

SELFBACK-Activity Recognition for Self-management of Low Back Pain
Sani, S., Wiratunga, N., Massie, S. and Cooper, K.
SGAI International Conference on Artificial Intelligence – SGAI 2016

2017

Learning Deep and Shallow Features for Human Activity Recognition
International Conference on Knowledge Science, Engineering and Management – KSEM 2017
knn Sampling for Personalised Human Activity Recognition
International Conference on Case-Based Reasoning – ICCBR 2017
Learning Deep Features for kNN-Based Human Activity Recognition
Workshop on Case-based Reasoning and Deep Learning – CBRDL 2017
A Siamese Convolutional Network for Developing Similarity Knowledge in the SelfBACK Dataset
Workshop on Case-based Reasoning and Deep Learning – CBRDL 2017
Accuracy of Physical Activity Recognition from a Wrist-worn Sensor
Physiotherapy UK 2017

2016

Data-Driven Self-Management of Chronic Health Conditions
UK Workshop on Case-Based Reasoning – UKCBR 2016

ntnu
hle
rgunfaTrade Expansion sduuog



Datasets
Tri-axial accelerometer data recorded on both the wrist and the thigh. Data is sampled at 100Hz and covers the activities of lying, sitting, standing, walking, walking up stairs, walking down stairs and running.
Code
Python code for activity recognition, Including deep convolutional neural network models.