A5087896268- Vestibular and auditory disorders
- Obstructive Sleep Apnea Research
- Context-Aware Activity Recognition Systems
- Balance, Gait, and Falls Prevention
- Indoor and Outdoor Localization Technologies
- Wireless Networks and Protocols
- Health disparities and outcomes
- Neuroscience and Music Perception
- Text and Document Classification Technologies
- Mind wandering and attention
- Ophthalmology and Visual Impairment Studies
- Topic Modeling
- Machine Learning and Algorithms
- Neural and Behavioral Psychology Studies
Ipswich Hospital
2023
University of Essex
2019
Insigneo
2017
University of Sheffield
2017
Newcastle University
2014
The study of spontaneous and everyday cognitions is an area rapidly growing interest. One the most ubiquitous forms cognition involuntary musical imagery (INMI), involuntarily retrieved repetitive mental replay music. present introduced a novel method for capturing temporal features INMI within naturalistic setting. This allowed investigation two questions interest to researchers in more objective way than previously possible, concerning (1) precision memory representations (2) interactions...
Falls have always been one of the major threats to health and well-being elderly people, particularly for those living alone. Both wearable non-wearable fall detection systems already developed. However, using WiFi channel state information (CSI) attracted a significant interest from researchers due their non-intrusive low-cost nature. There are existing machine learning (ML) based CSI; however, most trained with comprehensive datasets tend achieve relatively lower accuracy compared that...
Vestibular disorders affect an individual's stability, balance, and gait predispose them to falls. Traditional laboratory-based semi-objective vestibular assessments are intrusive cumbersome provide little information about their functional ability. Commercially available wearable inertial sensors allow us make this real life objective, with a detailed view of abilities. Timed Up Go (TUG) Postural Sway tests commonly used for balance assessments. Our aim was assess the feasibility,...
The Timed Up and Go (TUG) test is widely used for assessing mobility falls risk of elderly individuals. In this study, we aim to utilize TUG estimate disability level community dwelling participants. Forty features are extracted from single wrist mounted accelerometer signals which recorded in home environment the 321 participants performing test. As an initial exploratory analysis, linear discriminant classifier levels. study compares models built using with standard measure time taken...
Classification of short texts (e.g. reviews, sentences) is a well-defined task, usually attempted in regimes where data abundant. This is, however, not always the case. Limited availability very common industrial settings and seriously hinders performance any classification task-it obvious how to perform augmentation. In this work, we apply Recurrent Neural Network Monte Carlo Tree Search (MCTS) generate unlabelled questions. We use Human-In-the-Loop help decide whether 1) generated...