Project
AQUABRAIN
Unit(s) of assessment: Computer Science and Informatics
Research theme: Computational Intelligence and Applications Research Group (CIA)
School: School of Science and Technology
Overview
Traditionally, the observation and analysis of underwater species and ecosystems have relied on invasive or non-real-time methods, such as trapping or pre-recording using conventional cameras and microphones. Unfortunately, these approaches are limited in their ability to provide real-time insights, hindering the capacity for timely decision-making in underwater environments. Our impact case study is centred on a transformative goal – the development of AI-driven tools that empower users, including marine biologists, aquaculture farmers, and ecologists, to conduct real-time monitoring and decision-making related to underwater life, pollutant agents, and various species. This innovation seeks to enable evidence-based, real-time decision-making, addressing critical challenges in aquatic research and conservation.
Key Objectives:
- Develop fully functional prototypes to enable the piloting of underwater real-time monitoring by the conclusion of 2023. The ICS will provide the foundation for practical, real-time data collection and analysis beneath the water's surface.
- Collaborate with potential partners and users who have a proven track record in underwater monitoring, utilising traditional methods. The engagement will lead to the design of feasibility studies that are tailored to the unique demands of real-time monitoring, culminating by the end of 2024.
- Pursue external project opportunities alongside partners who have demonstrated success in securing funding in areas such as Sustainable Aquaculture, Marine Ecosystems, and Fish Health and Welfare. These partnerships will enhance the project's reach and its impact in the specified domains by the end of 2024.
- Forge partnerships with industrial experts to advance the prototype to Technology Readiness Levels (TRL) 7 to 8, signifying readiness for deployment and application. The collaborative effort will propel the project toward realisation by the end of 2025, ensuring that our AI-driven tools reach their full potential in the field of underwater real-time monitoring and decision-making.
Gallery
The AQUABRAIN impact case focused on Behavioural Analysis for AI-assisted Decision-Making (BEHAVIOR-AI) is underpinned by research focusing on the utilisation of Intelligent Edge devices, which operate independently of the internet using AI and Machine Learning algorithms. This research has generated valuable insights and findings by integrating behavioural analysis with AI algorithms, with applications in healthcare [R1, R5], agriculture/aquaculture [R1, R2, R3, R4] and environmental monitoring [R1, R2, R3, R4]. Personalised temperature regulation using Intelligent Edge devices has been identified as a significant factor in improving well-being and preventing heat-related illnesses.
Dr Pedro Machado and Dr Isibor Kennedy Ihianle have contributed their expertise to the research, with their respective interests in neuromorphic engineering, edge computing, and pervasive computing. The research is interdisciplinary, emphasising sustainable practices, and aligned with the Sustainable Development Goals. The contextual aspects include advancements in edge computing, interdisciplinary collaboration, and the aim to enhance health and quality of life while addressing climate action and the preservation of aquatic ecosystems.
The outputs and bodies of work still in development for the impact case study are maturing and undergoing refinement. Publication plans include research papers and technical reports that provide detailed insights into the methodology, findings, and implications of the research. These publications are expected to be released within the next 6 to 18 months. Some outputs will be co-authored with non-academic partners and stakeholders to ensure their direct applicability and relevance.
The research is closely aligned with the anticipated impact, focusing on improving health outcomes, enhancing agricultural practices, and promoting environmental sustainability. It leverages behavioural analysis and AI algorithms to make informed decisions, optimise resource usage, and improve overall well-being. The research operates at the intersection of various disciplines, such as computer science, healthcare, agriculture, and environmental science. It embraces advancements in edge computing, AI algorithms, and data analytics. Contextual information includes the emphasis on sustainability, climate action, and the preservation of aquatic ecosystems. The research promotes interdisciplinary collaboration and engages with non-academic stakeholders to ensure practical application and real-world impact.
The area of research related to AQUABRAIN involves several key contextual aspects that are relevant to understanding its significance and impact. Some key contextual information includes:
AQUABRAIN's development of advanced AI algorithms enables real-time and accurate classification of fish species, identification of gender and life stages, and precise weight estimation. These advancements enhance the capacity to monitor aquatic ecosystems dynamically, allowing for timely responses to changes in fish behaviour and ecosystem dynamics.
By integrating Intelligent Edge devices, such as the Intel RealSense Camera D435i, hydrophones, and LiDAR L515, into the monitoring system, the project addresses challenges related to underwater visibility, lighting conditions, and aquatic environments. This integration facilitates real-time data collection and analysis, providing a more comprehensive understanding of aquatic habitats.
A critical aspect of the research involves addressing potential ecological impacts associated with deploying monitoring systems in aquatic environments. The project's commitment to ethical considerations, permits, and guidelines ensures responsible deployment, minimising disruptions to aquatic ecosystems and avoiding harm to aquatic species.
The project gathers a comprehensive dataset of aquatic species interactions and behaviours using integrated sensors and AI algorithms. This dataset contributes to improved observation capabilities of biodiversity in UK/European waters, aiding in the conservation and management of aquatic ecosystems.
A unique contribution lies in the generation of a synthetic dataset using advanced AI techniques. This dataset enhances the detection of atypical behaviours associated with the early stages of fish disease outbreaks. Collaboration with experts in aquaculture, veterinary science, marine biology, and ecology strengthens the project's capacity for early detection and proactive mitigation of aquatic health threats.
AQUABRAIN's commitment to continuously optimising and refining the integrated system based on testing results and real-world data ensures adaptability to varying aquatic conditions. This ongoing refinement process aims to address challenges, improve accuracy, and enhance the overall reliability of the monitoring system.
The project actively promotes knowledge exchange and dissemination of findings through workshops, conferences, and publications. By engaging with industry partners, stakeholders, and the scientific community, AQUABRAIN contributes to innovation in environmental monitoring technologies and practices, offering a novel approach for addressing agricultural pollution and supporting conservation efforts.
AQUABRAIN enhances the skills and expertise of the research team by providing hands-on experience in developing and deploying cutting-edge technology for environmental monitoring. The interdisciplinary collaboration fosters technical innovation, combining expertise from computer science, environmental science, and aquaculture to address complex challenges from diverse perspectives.
A significant emphasis is placed on ensuring regulatory compliance and adherence to ethical guidelines throughout the research and development phases. The project aims to demonstrate responsible and ethical research practices, promoting transparency, and being accountable for the impact of the technology on aquatic environments and species.
Collaboration
The AQUABRAIN research program, in collaboration with esteemed partners, generates widespread benefits across key stakeholders. The Technical University of Mombasa (TUM) contributes scientific precision to aquaculture, ensuring tailored solutions for disease prevention and sustainable practices. Mediprospects AI offers cutting-edge AI tools, enhancing data analysis and operational efficiency.
The FishEthosGroup focuses on welfare and stakeholder collaboration, bridging gaps in the aquaculture sector. Collaborating with experts from the University of Stirling, University of Saint Andrews, and INESCTEC enriches the project with aquatic health modelling, microbiology insights, and formal methods expertise. AgriEpiCentre, with Charlie Bowyer, brings agricultural experience for precision practices.
Stemmer Imaging, through Robert Hutton-Attenborough, ensures seamless IoW sensor technology development, promising innovation in machine vision applications. This collaborative effort advances aquaculture practices, environmental sustainability, and industry efficiency on a global scale.
The TUM has a strong track record in developing scientific approaches for precision aquaculture, contributing through research and development, technical expertise, collaboration, knowledge exchange, and capacity building. They conduct studies, analyse data, and provide tailored solutions for disease prevention, environmental monitoring, and sustainable practices. Their partnership promotes knowledge exchange, engages stakeholders, and empowers the aquaculture sector through training and educational programs, aiming to improve standards and enhance sustainability in aquaculture operations.
Mediprospects AI specialises in developing commercial AI solutions for precision aquaculture, offering AI-based tools to analyse aquaculture data and improve practices. They focus on technical implementation, working closely with stakeholders to address specific challenges. Commercialisation and scalability are emphasised, enabling wider adoption of AI in aquaculture. Ongoing improvement and support ensure the solutions remain effective and up-to-date. Mediprospects AI's contributions enhance efficiency and sustainability in aquaculture operations, promoting the use of AI technologies across the industry.
The FishEthosGroup, a non-profit association, focuses on aquaculture welfare, aiming to bridge gaps between science and industry stakeholders, including producers, certifiers, retailers, NGOs, policymakers, and consumers. Founded in 2018 as a research group under the fair-fish association, it became an independent entity. Based in Portugal, it collaborates on projects with the Centre of Marine Sciences (CCMAR).
Collaboration with Dr. Darren Green and Dr. Andrew Desbois enriches the AQUABRAIN project with extensive expertise. Dr. Green's background in aquatic health modelling, disease spread, and nutrition modelling brings valuable insights, while Dr. Desbois' expertise in microbiology aligns seamlessly with project objectives, offering innovative approaches and ethical considerations for effective microbial infection studies and host protection strategies.
Collaboration with Prof. Juliana Bowles brings expertise in advanced formal methods to healthcare. Her focus on medication safety, showcased in EPSRC-supported projects, extends to predictive cancer treatment analysis for elderly individuals with comorbidities. Prof. Bowles' commitment to enhancing healthcare efficiency, exemplified in a Royal Academy of Engineering project, aligns with her contributions to developing advanced AI decision support systems, benefitting clinical processes and fish well-being.
Dr. Filipe Neves dos Santos, a seasoned Senior Research Coordinator at INESCTEC, has actively contributed to over 40 projects in the realms of agriculture and underwater robotics. His extensive experience includes serving as a reviewer for EU Horizon and H2020 projects. Engaging with Dr. Santos and INESCTEC not only leverages his wealth of expertise but significantly enhances the likelihood of successfully securing Horizon funding through collaborative initiatives.
Collaboration with Charlie Bowyer significantly enhances the Agri-EPI Centre by bringing a unique blend of agricultural expertise and industry experience. With a background in mixed farming and hands-on involvement in the biogas sector, Charlie's role as Business Development Manager for Livestock and Aquaculture focuses on driving precision agriculture research. His commitment to innovation and data leverage promises impactful advancements in agri-tech.
Collaboration with Robert Hutton-Attenborough for technology transfer to STEMMER IMAGING leverages his extensive engineering knowledge and the company's status as a leading systems house for machine vision technology. This collaboration ensures a comprehensive range of services for industrial and non-industrial applications, promising synergy to advance IoW sensor technology and drive innovation in machine vision applications.
Staff
Dr Pedro Machado - Pedro’s expertise includes Neuromorphic Engineering, Edge Computer Vision, Bio-inspired Computing, robotics and Intelligent Sensors. Furthermore, Pedro was the PI of the Field Companion project, Grant agreement 600359 funded by InnovateUK that aimed at developing a smart pulveriser by combining Computer Vision, Edge-Computing and state-of-the-art Deep Learning algorithms. Pedro will focus on the project management and the development of intelligent IoT devices (edge devices) to reduce the impact of low visibility underwater, and interoperability between edge devices and the cloud.
Dr Isibor Kennedy Ihianle - Kennedy’s research interest includes applications of computational intelligence techniques and expertise in machine and deep learning techniques for object detection and tracking, human activity recognition, data management, ontology, data analysis and analytics. He will be responsible for data governance and adhere to data standards to facilitate data fusion.
Dr Farhad Fassihi - Farhad’s Research interests include systematic innovation and sustainability. Farhard has extensive experience in developing digital information systems and will be responsible to look at sustainability and factors concerning the short to mid-term issues which will have an impact on the outcomes and potential future development/deployment.
Dr Jordan Bird’ - Jordan's research interests involve Human-Robot Interaction, Artificial Intelligence, Machine and Deep Learning, Transfer Learning, and Data Augmentation. His main goal is to apply technology to aid in everyday work and help to improve real-world situations. Jordan will focus on how machine learning can be used to automate underwater observations.
Dr Salisu Yahaya - a Lecturer with the Department of Computer Science, where he is also a member of the Computational Intelligence and Applications (CIA) research group. His research interest is in the application of computational intelligence for human activity recognition, behaviour modelling, abnormality detection and pattern recognition.
Dr Doratha Vinkemeier - Doratha's research interest is in facial expression units. She will provide vital support to researchers in terms of statistical analysis using AI/ML.
Dr Niki Khan - Niki's research interest is in assessing stress in lobsters. She will play a key role in fish welfare and assessing fish stress.
Mr. Dennis Monari (PhD candidate) - Research focuses on fish interaction with the ecosystem using Vision Perception.
Mr. Feliciano Domingos (PhD candidate) - Research focuses on fish interaction with the ecosystem using: Audio Perception.
Volodymyr Ivanov (Research Assistant) - Responsible for manufacturing and testing the 6 prototypes
Dr Pedro Machado
Dr Machado is a Senior Lecturer in Computer Science and Course Leader for MSc Artificial Intelligence. His research interest is on Neuromorphic Engineering, Parallel, Heterogeneous and Distributed Computing
Dr Isibor Kennedy Ihianle
Dr Isibor Kennedy Ihianle is a Lecturer at the Department of Computing and Technology, Nottingham Trent University, United Kingdom. He is also a Fellow Higher Education Academy and a Member of the British Computing Society.
Farhad Fassihi-Tash
Farhad Fassihi-Tash
Jordan J. Bird
Dr. Jordan J. Bird is a Senior Lecturer in Computer Science at Nottingham Trent University, UK.
Salisu Yahaya
Salisu Yahaya
Doratha Vinkemeier
Doratha is a Lecturer in the Department of Computer Science.
Niki Khan
Dr. Niki Khan lectures on the general topics of animal physiology, welfare, and genetics. She is the Module Leader for BSc Animal Biology and BSc Agriculture.
Objectives
The AQUABRAIN addresses challenges in aquaculture and environmental monitoring through the integration of behavioural analysis and AI using Intelligent Edge devices. It enables personalised interventions, real-time monitoring, and optimised practices to improve aquaculture practices, prevent disease outbreaks, and promote environmental stewardship. Informed decision-making, resource optimisation, and reduced environmental impacts drive positive change, leading to sustainable aquaculture practices and increased resilience.
The research aims to enhance aquaculture practices, support ecosystem preservation, and generate positive impacts for the industry and the environment. The key objectives are:
Improve health outcomes in fish by enabling personalised interventions, real-time monitoring, and precision aquaculture through the integration of behavioural analysis and AI.
Foster sustainable aquaculture by integrating behavioural analysis and AI to optimise practices, increase yields, minimise resource usage, and mitigate environmental impacts for enhanced food security and resilience.
Enhance environmental stewardship by preventing disease outbreaks, preserving aquatic ecosystems, and minimising environmental degradation, with a focus on early disease detection, promoting eco-friendly practices, and supporting the preservation of life below water.
Funding
- NTU: 23/24 IAF-funded ICS
- NTU: CIRC Quality-Related Research funding
Publications
R1. MURRAY-HILL, N., FONTES, L., MACHADO, P. and IHIANLE, I.K., 2023. Secure video streaming using dedicated hardware. Journal of Signal Processing Systems. ISSN 1939-8018 - – Predicted: 3*
R2. MACHADO, P., FERREIRA, J.F., OIKONOMOU, A. and MCGINNITY, T.M., 2023. NeuroHSMD: neuromorphic hybrid spiking motion detector. ACM Transactions on Reconfigurable Technology and Systems. ISSN 1936-7406 – Predicted: 4*
R3. MAGALHÃES, S.C., SANTOS, F.N., MACHADO, P., MOREIRA, A.P. and DIAS, J., 2023. Benchmarking edge computing devices for grape bunches and trunks detection using accelerated object detection single shot multibox deep learning models. Engineering Applications of Artificial Intelligence, 117 (Part A): 105604. ISSN 0952-1976 – Predicted: 4*
R4. MACHADO, P., OIKONOMOU, A., FERREIRA, J.F. and MCGINNITY, T.M., 2021. HSMD: an object motion detection algorithm using a Hybrid Spiking Neural Network Architecture. IEEE Access. ISSN 2169-3536 – 3*
R5. Ihianle, I.K., Nwajana, A.O., Ebenuwa, S.H., Otuka, R.I., Owa, K. and Orisatoki, M.O., 2020. A deep learning approach for human activities recognition from multimodal sensing devices. IEEE Access, 8, pp. 179028-179038. ISSN 2169-3536 – Predicted: 3*
R6 Dennis Monari, Jack Larkin, Pedro Machado, Jordan J. Bird, Isibor Kennedy Ihianle, Salisu Wada Yahaya, Farhad Fassihi Tash, Md Mahmudul Hasan, Ahmad Lotfi; UDEEP: Edge-based Computer Vision for In-Situ Underwater Crayfish and Plastic Detection. Elsevier Internet of Things, submitted – Predicted: 4*