Hamzah S. AlZu'bi

Smart Bolus Estimation Taking into Account the Amount of Insulin on Board

Self-management blood glucose (SMBG) and bolus calculations are pivotal components of evidence-based standard of care for young diabetics receiving multiple daily insulin injections. This paper aims at developing a smart bolus estimator that takes into account the amount of insulin on board (IoB), i.e. Insulin remaining in the patients body, to reduce fear of hypoglycemia and achieve goals of glycemic control. Design of the proposed bolus estimator follows the feed-forward multi-perceptron artificial neural network (ANN) with an input represents the time shift from last injection of quick acting (QA) insulin. The network output represents the amount of IoB. Calculation of the bolus required for a meal/snack is then carried out normally deducing the estimated IoB and thus the possibility of excess insulin administration and the risk of hypoglycemia can be averted. A functional prototype of the proposed system is developed and tested successfully on various mobile devices (i.e smartphones and tablet computers).

Toward Inexpensive and Practical Brain Computer Interface

Brain Computer Interface (BCI) introduces a new communication system that does not depend on brains normal output pathways. This paper studies the feasibility of using inexpensive Electroencephalogram(EEG) device for BCI with asynchronous BCI mode which leads the control mechanism to become highly available for variety of users and a more natural way to communicate than the current BCI versions which depend on expensive EEG devices, which are less practical where they need special environment to run the BCI system, and use synchronous BCI mode in most cases. Benchmarking the results using Emotiv to another complex EEG device (BrainAmp) shows the reliability of such inexpensive devices.

Prediction of driver fatigue: Approaches and open challenges

Fatigue is a mental process that grows gradually and affects human reaction time and the consciousness. It is one of the causes of road fatal accidents around the globe. Although it is now generally accepted that fatigue plays an important role in road safety, it is still largely left to individual drivers to manage. The recent research in this area focuses on fatigue detection and the existing systems alert the drivers in severe fatigued stage. These systems use either physiological signs of the fatigue or the behavioural reaction to generate alerts. This research investigates the feasibility of using a group of fatigue symptoms (such as pupil response, gaze patterns, steering reaction and EEG) to build a robust fatigue detection algorithm that can be used in a real-life system for the early prediction and avoidance of fatigue development. Intensive testing and validation stages are required to ensure the reliability and the suitability of the system that should be able to detect fatigue levels at different degrees of tiredness. Moreover, the proposed system predicts subsequent stages of fatigue and generates an approximate behavioural model for each individual driver to enable more personalised and effective intervention.

Human activity classification using a single accelerometer

Human activity recognition is an area of growing interest facilitated by the current revolution in body-worn sensors. Activity recognition allows applications to construct activity profiles for each subject which could be used effectively for healthcare and safety applications. Automated human activity recognition systems face several challenges such as number of sensors, sensor precision, gait style differences, and others. This work proposes a machine learning system to automatically recognise human activities based on a single body-worn accelerometer. The in-house collected dataset contains 3D acceleration of 50 subjects performing 10 different activities. The dataset was produced to ensure robustness and prevent subject-biased results. The feature vector is derived from simple statistical features. The proposed method benefits from RGB-to-YIQ colour space transform as kernel to transform the feature vector into more discriminable features. The classification technique is based on an adaptive boosting ensemble classifier. The proposed system shows consistent classification performance up to 95% accuracy among the 50 subjects.

Estimation of driver alertness using low-cost wearable devices

Operator tiredness and fatigue pose significant safety risks in industries such as mining, trucking, aviation and air traffic control. Fatigue is a major factor behind on-the-job accidents, absenteeism and lowered productivity. The detection and prediction of fatigue is a technical challenge, confounded by the practical constraints of the working environment. This paper presents a mathematical fatigue model using low cost wearable biosensors, customized for each individual participant. The circadian rhythm mathematical model is the major part of fatigue model core. Heart rate and skin temperature, which are known circadian pacemakers, are used to build the proposed mathematical model. The customization and adaptation of the circadian is expected to improve the fatigue model and make it more accurate in the context of individuality and adaptation to the time of day and the activity at hand. Using low-cost, non-intrusive and portable biosensors allows the model to be tested in real working conditions with minimal impact on the operators. In addition to presenting a mathematical model for the customized circadian, this paper presents some preliminary results from a limited experiment, with some promising results.

Towards automated monitoring of adult zebrafish

Over the last two decades, zebrafish (Danio rerio) have emerged as an efficient model to aid in the research of a broad range of human diseases as well as such diverse applications as environmental modelling and drug discovery. Economically, the large number, low price and low maintenance requirements of this fish species encouraged its use for research. In addition to this, the study of zebrafish is being used to improve the understanding of fish physiology, with implications for fish welfare. In order to thoroughly model the behaviour, development and growth of these fish, it is important to be able to scrutinise the characteristics of individual fish as they respond to a range of stimuli, and to this end off-line fish recognition and on-line tracking using video data is employed. Tracking and identifying such small and fast-moving objects is a challenge, and this paper seeks to address this using a behavioural analysis approach. Utilising single high resolution camera and two low-cost synchronised video cameras, the proposed systems captures front (face) and side (profile) pictures of each isolated fish as they swim past a given marker. The acquired images are then subject to three separate processing routes in order to satisfy three complementary but distinct objectives. Initially, fish face and profile features are extracted to aid the identification of individual fish. Then, for each fish identified, behavioural features such as the frequency and intensity of the operculum beat rate or breathing cycle are quantified in order to assess aspects of the fish welfare. Additionally, the volume of each fish is estimated based on its profile dimensions, enabling the weight of the fish to be monitored throughout its lifetime. This paper presents preliminary experimental considerations and findings of this on-going research project. Results to date have been both encouraging and promising, validating the approach and the experimental configuration adopted.

Horse stress analysis using biomechanical modelling and machine learning approach

Horse transport is a common practice in the equestrian industry, especially with the expansion of this industry around the world. Research has proved that horse transport by road is responsible for high stress levels, which sometimes exceed stress levels caused by exercising during professional horse races. Stress symptoms are reflected in the physiological functions of horses leading to horses suffering from horses fatigue or the injury. The horses stand still in a small box during transport to ensure safety and avoid falls or injuries. The weight is held by the four limbs while the vehicle is moving and vibration forces keep interrupting the balance. This requires the horse to counter these forces in order to keep its balance which demands high energy consumption even for short trips. The horse blood circulation system tries to support the muscles with enough oxygen forcing the heart to beat at high rates. This paper suggests an analytical biomechanical model for the vibration forces to understand how these forces move through horse limbs. This model is proposed to associate vibration forces with high stress levels during transport. Such a direct relationship between vehicle vibration forces and high stress levels will lead to a low cost non-invasive early stress detection system without the need to measure any direct physiological response of the horse. This relationship will also shed light on the importance of optimised vehicle design to reduce vibrations.

An intelligent behavior-based fish feeding system

Aquaculture is a growing multi-billion pound industry facing many challenges. Traditional fish feeding mechanism in todays aquaculture farms stands behind a variety of challenges, including fish welfare, fish growth distribution, environmental effect especially in open ocean cage fish farms, and production cost efficiency. Adaptive smart fish feeder based on fish behaviours is proposed in this paper in order to minimize the effect of the traditional feeding mechanisms. The proposed feeding mechanism interacts, recognizes and responses to fish activities. The proposed smart fish feeder feeds fish based on their request regardless the time of the day. The smart fish feeding aims to minimize food waste and maximize the food conversion ratio (FCR). The proposed system is expected to cause uniform fish growth among individuals within the tank as the feeding depends on fish requests. Fish welfare is expected to be enhanced since there is no food competition and food waste is expected to be less making water good quality last for longer. This paper proposes hardware design of the smart feeder and smart software algorithm. Preliminary results will be discussed in this paper.

Real-time 3D fish tracking and behaviour analysis

Over half a million fish are used in scientific procedures annually in the UK alone. Most fish are subject to invasive procedures which may cause pain distress or death. A range of procedures such as fin clipping, tagging and exposure to chemicals of low pH have been associated with change in behaviour. Abnormal behaviour after common procedures may influence and confuse experiment output.