Ryan Cunningham

Ryan Cunningham
Manchester Metropolitan University | MMU · Department of Computing and Mathematics

Doctor of Philosophy

About

31
Publications
6,605
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
171
Citations
Citations since 2016
27 Research Items
167 Citations
2016201720182019202020212022010203040
2016201720182019202020212022010203040
2016201720182019202020212022010203040
2016201720182019202020212022010203040
Introduction
Part of the Department of Computing and Mathematics at Manchester Metropolitan University, teaching and researching in Programming Languages, HPC, Deep Learning and Image Analysis.
Additional affiliations
January 2019 - present
Manchester Metropolitan University
Position
  • Lecturer
Description
  • Teaching: (1) TensorFlow for HPC/Deep Learning (MSc) (2) Programming Languages: Principles and Design (Compilers) (BSc)
February 2015 - February 2019
Manchester Metropolitan University
Position
  • Research Associate
Description
  • Deep learning medical imaging solution.
Education
January 2012 - January 2015
Manchester Metropolitan University
Field of study
  • Computing
September 2009 - May 2011
Manchester Metropolitan University
Field of study
  • Artificial Intelligence

Publications

Publications (31)
Article
Full-text available
Despite widespread availability of ultrasound and a need for personalised muscle diagnosis (neck/back pain-injury, work related disorder, myopathies, neuropathies), robust, online segmentation of muscles within complex groups remains unsolved by existing methods. For example, Cervical Dystonia (CD) is a prevalent neurological condition causing pain...
Preprint
Full-text available
This paper presents an investigation into the feasibility of using deep learning methods for developing arbitrary full spatial resolution regression analysis of B-mode ultrasound images of human skeletal muscle. In this study we focus on full spatial analysis of muscle fibre orientation, since there is an existing body of work with which to compare...
Article
Full-text available
The aim of this study was to provide automated identification of postural point-features required to estimate the location and orientation of the head, multi-segmented trunk and arms from videos of the clinical test 'Segmental Assessment of Trunk Control' (SATCo). Three expert operators manually annotated 13 point-features in every fourth image of...
Article
Full-text available
Objective: To provide objective visualization and pattern analysis of neck muscle boundaries to inform and monitor treatment of cervical dystonia. Methods: We recorded transverse cervical ultrasound (US) images and whole-body motion analysis of sixty-one standing participants (35 cervical dystonia, 26 age matched controls). We manually annotated...
Article
Full-text available
The objective is to test automated in vivo estimation of active and passive skeletal muscle states using ultrasonic imaging. Current technology (electromyography, dynamometry, shear wave imaging) provides no general, non-invasive method for online estimation of skeletal muscle states. Ultrasound (US) allows non-invasive imaging of muscle, yet curre...
Article
Full-text available
Long video datasets of facial macro- and micro-expressions remains in strong demand with the current dominance of data-hungry deep learning methods. There are limited methods of generating long videos which contain micro-expressions. Moreover, there is a lack of performance metrics to quantify the generated data. To address the research gaps, we in...
Preprint
Micro-expression spotting is the preliminary step for any micro-expression related analysis to avoid excessive false positives. We propose an efficient lightweight macro- and micro-expression spotting method which takes advantage of the duration differences of macro- and micro-expressions. Using effective frame skips, local contrast normalisation,...
Preprint
Full-text available
Objective: To test automated in vivo estimation of active and passive skeletal muscle states using ultrasonic imaging. Background: Current technology (electromyography, dynamometry, shear wave imaging) provides no general, non-invasive method for online estimation of skeletal intramuscular states. Ultrasound (US) allows non-invasive imaging of musc...
Preprint
Full-text available
Objective: To provide an automated visualization, pattern analysis and classification of neck muscle boundaries comparing cervical dystonia with healthy controls. Methods: We recorded transverse cervical ultrasound (US) images and whole-body motion analysis of sixty-one standing participants (35 cervical dystonia, 26 age matched controls). We manua...
Preprint
Full-text available
Objectives: To automate online segmentation of cervical muscles from transverse ultrasound (US) images of the human neck during functional head movement. To extend ground-truth labelling methodology beyond dependence upon MRI imaging of static head positions required for application to participants with involuntary movement disorders. Method: We co...
Preprint
Full-text available
Objective: To automate identification of postural point-features from colour videos of children with neuromuscular disorders, during clinical assessment. The automatic identification of 13 points of interest (2, 6, 2, 3 points on the head, trunk, pelvis, arm respectively) is required to estimate the location and orientation of head, trunk, and arm...
Data
Instructions for accessing the data set used in the paper "Estimating full regional skeletal muscle fibre orientation from B-mode ultrasound images using convolutional, residual, and deconvolutional neural networks". We have provided a .mat file which opens with Matlab. Once opened, you should see two items in the workspace: 1) data a cell array...
Preprint
Full-text available
Background The Segmental Assessment of Trunk Control (SATCo) evaluates sitting control at seven separate trunk segments, making a judgement based on their position in space relative to a defined, aligned posture. SATCo is in regular clinical and research use and is a Recommended Instrument for Cerebral Palsy and Spinal Cord Injury-Paediatric by The...
Article
Full-text available
This paper presents an investigation into the feasibility of using deep learning methods for developing arbitrary full spatial resolution regression analysis of B-mode ultrasound images of human skeletal muscle. In this study, we focus on full spatial analysis of muscle fibre orientation, since there is an existing body of work with which to compar...
Preprint
Direct measurement of strain within muscle is important for understanding muscle function in health and disease. Current technology (kinematics, dynamometry, electromyography) provides limited ability to measure strain within muscle. Regional fiber orientation and length are related with active/passive strain within muscle. Currently, ultrasound im...
Preprint
Direct measurement of strain within muscle is important for understanding muscle function in health and disease. Current technology (kinematics, dynamometry, electromyography) provides limited ability to measure strain within muscle. Regional fiber orientation and length are related with active/passive strain within muscle. Currently, ultrasound im...
Preprint
Full-text available
This paper concerns the fully automatic direct in vivo measurement of active and passive dynamic skeletal muscle states using ultrasound imaging. Despite the long standing medical need (myopathies, neuropathies, pain, injury, ageing), currently technology (electromyography, dynamometry, shear wave imaging) provides no general, non-invasive method f...
Conference Paper
Full-text available
This paper concerns fully automatic and objective measurement of human skeletal muscle fiber orientation directly from standard b-mode ultrasound images using deep residual (ResNet) and convolutional neural networks (CNN). Fiber orientation and length is related with active and passive states of force production within muscle. There is currently no...
Article
Full-text available
While individual muscle function is known, the sensory and motor value of muscles within the whole-body sensorimotor network is complicated. Specifically, the relationship between neck muscle action and distal muscle synergies is unknown. This work demonstrates a causal relationship between regulation of the neck muscles and global motor control. S...
Conference Paper
Multivariable intermittent control (MIC) combines stability with flexibility in the control of unstable systems. Using an underlying continuous-time optimal control design, MIC uses models of the physical system to generate multivariate open-loop control signals between samples of the observed state. Using accurate model values of physical system p...
Thesis
Full-text available
Skeletal muscles control the joints of the skeletal system and they allow human movement and interaction with the environment. They are vital for stability in balance, walking and running, and many other skilled motor tasks. To understand how muscles operate in general and specific situations there are a variety of tools at the disposal of research...
Conference Paper
Full-text available
Traditionally surface Electromyography (EMG) has been used to non-invasively identify the activation of superficial skeletal muscle. We propose that the automated analysis of Ultrasound (US) images can provide an alternative technique by which active and passive muscle movement may be classified. We present a method by which the change in muscle sh...
Conference Paper
Full-text available
The aim of this paper is to track objects during their use by humans. The task is difficult because these ob-jects are small, fast-moving and often occluded by the user. We present a novel solution based on cascade action recognition, a learned mapping between body-and object-poses, and a hierarchical extension of im-portance sampling. During track...
Conference Paper
Full-text available
The presence of involuntary muscle twitches is a di-agnostic indicator of neurodegenerative diseases, such as motor neurone disease (MND), but current methods of twitch detection are invasive and pose potential risks to patients. We present a method by which standard B-mode ultrasound can be used to automatically iden-tify muscle twitches similar t...

Network

Cited By

Projects

Projects (5)
Project
Introduction Everyday activities functional activities such as looking around, sitting or any use of our arms and hands, relies on our having good control of our head and trunk. But children who have a neuromotor disability such as cerebral palsy (CP), or a neuromuscular disorder (NMD) such as spinal muscular atrophy (SMA) or one of the other inherited NMDs, often experience major difficulties with effective trunk control. This inevitably impacts on their functional abilities. Research has shown a direct relationship between poor trunk control (orienting the head and sitting ability) in children with CP: this can limit their use of eye gaze systems (a way of accessing a communication aid using a mouse that you control with your eyes), can make it difficult to use both hands simultaneously for learning, play and social development as one hand is needed to maintain balance, and can affect their mobility and urinary continence. For children with more severe CP, these functional abilities decline even further from around age 7 years. But research has also shown that accurate diagnosis of trunk control problems can enable a tailored physiotherapy programme which, when combined with treatment as usual, can improve functional abilities, even for those children with severe CP. What is needed is an accurate, objective tool to measure trunk control status that is suitable for use by different therapists and healthcare professionals in their own centres and clinics. Infants with SMA Type 1, the severest form of SMA, have similar difficulties in learning trunk control to children with severe CP. We need to be able to measure the detailed effect on trunk control of the new drug Nusinersen, and other emerging drug treatments, so that we can better understand their impact and monitor any changes over time. For children with other Neuromuscular Disorders (such as SMA Type2, congenital myopathy, congenital muscular dystrophy, Duchenne muscular dystrophy), their increasing muscle weakness leads to loss of trunk control that they may have gained: for many children, this can lead to curving of the spine (scoliosis). Objective measurement, sensitive enough to detect changes in trunk control, pre-scoliosis, would mean that appropriate treatment, such as physiotherapy, could be given at the best possible time. What assessment tools are available now? All current assessments of trunk control suitable for use in a clinic are clinically agreed and standardised but rely on therapists’ judgement (opinion) that must be maintained consistently over time and between therapists. These assessments usually consider ‘the trunk’ as a single entity and thus lose the accurate and detailed assessment of trunk sub-divisions (or trunk segments) that is needed for individualised tailored therapy. Many of these assessments can only be used if a child is already able to sit by themselves and are not appropriate for children with severe disability. Our Proposal Our proposal is to develop an assessment tool to quantify trunk control on a segmental basis, and that is suitable for use in a routine physiotherapy clinic without changing normal routine beyond setting up two cameras and a linked computer. Our hypothesis is that this objective measurement, classified to seven segmental levels, would bring benefits in terms of: • new understanding of how trunk control is gained and is maintained • monitoring treatment over time • providing evidence for all therapists who aim to improve trunk and postural control and associated functional abilities: in turn, this would improve therapy practice generally. In summary, this new assessment tool could lead to new interventions to enhance trunk control, and thus functional skills, for thousands of children worldwide, including those with greater severity of CP. Our project will: i) develop live imaging analysis technology to automate the assessment of trunk control in sitting ii) provide a cost effective and accessible tool usable within any clinic without increasing assessment time and iii) publish an online interactive database disseminating a public standard, a reference and a training resource for measuring trunk control, increasing understanding and enhancing expertise. Who we will enrol? Assessments of trunk control will be made on: • 400 children with CP, • 20 children with SMA Type 1 • 40 typically developing children (TD) • 40 children with other NMD Assessments will be recorded using standard 3D cameras available for the home market and each child will be seen just once for this assessment. The dataset we will gather will include examples of all trunk control problems, body shapes and sizes and different heights that the clinical tool would be expected to meet in the future. Machine learning methods will be used to train and confirm (validate) the automated diagnosis. The tool will be validated against expert clinical assessment and be made available as a laptop, software and two cameras, suitable for use by clinical staff in a routine clinical environment. We will also collect a longitudinal dataset of 20 children with CP and 5 infants with SMA 1 during a twelve-month period when each child will be seen at two month intervals. This will enable us to prepare statistical calculations necessary to plan clinical trials that test different treatments against each other. This work is ongoing at Manchester Metropolitan University, under the leadership of Professor Ian Loram. It is funded by the Medical Research Council.
Project
Assess the use of ultrasound of neck muscles using deep learning, for diagnostic, functional and therapeutic information.