Generalisation challenges in deep learning models for medical imagery: insights from external validation of COVID-19 classifiers.

Haynes, Sophie Crawford; Johnston, Pamela; Elyan, Eyad

doi:10.1007/s11042-024-18543-y

An evolutionary neural architecture search-based approach for time series forecasting. (2025)
Presentation / Conference Contribution
VU, T.H., NGUYEN, T.T. and ELYAN, E. 2025. An evolutionary neural architecture search-based approach for time series forecasting. In Proceedings of the 2025 IEEE (Institute of Electrical and Electronics Engineers) Congress on evolutionary computation (CEC 2025), 8-12 June 2025, Hangzhou, China. Piscataway: IEEE [online], article number 11043002, pages 1-8. Available from: https://doi.org/10.1109/cec65147.2025.11043002

Time series forecasting (TSF) is one of the most prevalent research topics in artificial intelligence and has garnered significant attention in the research community. In recent years, significant breakthroughs have been made in TSF research, shiftin... Read More about An evolutionary neural architecture search-based approach for time series forecasting..

CEANet: an end-to-end transformer based static-mobile points alignment approach. (2025)
Journal Article
YESIPOV, V., MEKALA, M.S., EDGE, P. and ELYAN, E. 2025. CEANet: an end-to-end transformer based static-mobile points alignment approach. Signal, image and video processing [online], 19(8), article number 596. Available from: https://doi.org/10.1007/s11760-025-04258-6

Correspondence-based Point Cloud Registration (PCR) is crucial for 3D visualization applications, especially in change detection. Most PCR models depend on precise initialization using a set of closest points to establish correspondences, a method th... Read More about CEANet: an end-to-end transformer based static-mobile points alignment approach..

Predicting student next-term performance in degree programs using AI-based approach: a case study from Ghana. (2025)
Journal Article
DIEKUU, J.-B., MEKALA, M.S., ABONIE, U.S., ISAACS, J. and ELYAN, E. 2025. Predicting student next-term performance in degree programs using AI-based approach: a case study from Ghana. Cogent education [online], 12(1), article number 2481000. Available from: https://doi.org/10.1080/2331186X.2025.2481000

Student performance can fluctuate over time due to various factors (e.g. previous assignment grades, social life and economic conditions). Temporal dynamics, such as semester-to-semester variations, and changes in students' academic achievements, beh... Read More about Predicting student next-term performance in degree programs using AI-based approach: a case study from Ghana..

Geothermal cooling solutions for rural communities at Homa Bay, Kenya: a CFD modelling study. (2025)
Journal Article
HOSSAIN, M., LASSALE, M., VERTIGANS, S., DEVECI, G., ELYAN, E., BURGESS, K. and OKOWA, M. 2025. Geothermal cooling solutions for rural communities at Homa Bay, Kenya: a CFD modelling study. International journal of sustainable engineering [online], 18(1), article number 2480095. Available from: https://doi.org/10.1080/19397038.2025.2480095

A Computational Fluid Dynamics (CFD) modelling study has been presented to analyse the cooling impact of a geothermal cooling system based on the Earth-Air-Tunnel-Heat-Exchanger (EATHE) concept to provide passive cooling to heat-stressed vulnerable p... Read More about Geothermal cooling solutions for rural communities at Homa Bay, Kenya: a CFD modelling study..

Blind sonar image quality assessment via machine learning: leveraging micro- and macro-scale texture and contour features in the wavelet domain. (2024)
Journal Article
TOLIE, H.F., REN, J., CHEN, R., ZHAO, H. and ELYAN, E. 2025. Blind sonar image quality assessment via machine learning: leveraging micro- and macro-scale texture and contour features in the wavelet domain. Engineering applications of artificial intelligence [online], 141, article number 109730. Available from: https://doi.org/10.1016/j.engappai.2024.109730

In subsea environments, sound navigation and ranging (SONAR) images are widely used for exploring and monitoring infrastructures due to their robustness and insensitivity to low-light conditions. However, their quality can degrade during acquisition... Read More about Blind sonar image quality assessment via machine learning: leveraging micro- and macro-scale texture and contour features in the wavelet domain..

A novel ensemble aggregation method based on deep learning representation. (2024)
Presentation / Conference Contribution
NGUYEN, T.T., ELYAN, E., DANG, T., NGUYEN, T.T. and LONGMUIR, M. 2025. A novel ensemble aggregation method based on deep learning representation. In Antonacopoulos, A., Chaudhuri, S., Chellappa, R., et al. (eds.) Pattern recognition: proceedings of the 27th International conference on pattern recognition, 1-5 December 2024, Kolkata, India. Lecture notes in computer science, 15324. Cham: Springer [online], pages 31-46. Available from: https://doi.org/10.1007/978-3-031-78383-8_3

We propose a novel ensemble aggregation method by using a deep learning-based representation approach. Specifically, we applied the Cross-Validation procedure on training data with a number of learning algorithms to obtain the predictions for trainin... Read More about A novel ensemble aggregation method based on deep learning representation..

Event classification on subsea pipeline inspection data using an ensemble of deep learning classifiers. (2024)
Journal Article
DANG, T., NGUYEN, T.T., LIEW, A.W.-C. and ELYAN, E. 2025. Event classification on subsea pipeline inspection data using an ensemble of deep learning classifiers. Cognitive computation [online], 17(1), article 10. Available from: https://doi.org/10.1007/s12559-024-10377-y

Subsea pipelines are the backbone of the modern oil and gas industry, transporting a total of 28% of global oil production. Due to several factors, such as corrosion or deformations, the pipelines might degrade over time, which might lead to serious... Read More about Event classification on subsea pipeline inspection data using an ensemble of deep learning classifiers..

A review of well life cycle integrity challenges in the oil and gas industry and its implications for sustained casing pressure (SCP). (2024)
Journal Article
IBUKUN, M., ELYAN, E., AMISH, M., NJUGUNA, J. and OLUYEMI, G.F. 2024. A review of well life cycle integrity challenges in the oil and gas industry and its implications for sustained casing pressure (SCP). Energies [online], 17(22), article number 5562. Available from: https://doi.org/10.3390/en17225562

Sustained Casing Pressure (SCP) is a condition in oil and gas wells where continuous pressure buildup in the well casing over a long period of time occurs. Several factors might be responsible for this, including the influx of formation fluids, the l... Read More about A review of well life cycle integrity challenges in the oil and gas industry and its implications for sustained casing pressure (SCP)..

Deep learning for digitising complex engineering drawings. (2024)
Thesis
JAMIESON, L. 2024. Deep learning for digitising complex engineering drawings. Robert Gordon University, PhD thesis. Hosted on OpenAIR [online]. Available from: https://doi.org/10.48526/rgu-wt-2795656

Vast amounts of documents are still commonly stored in undigitised formats. Consequently, the data they contain cannot be used to its full potential, as substantial manual effort is required to analyse it. Amongst these documents, engineering drawing... Read More about Deep learning for digitising complex engineering drawings..

Few-shot symbol detection in engineering drawings. (2024)
Journal Article
JAMIESON, L., ELYAN, E. and MORENO-GARCÍA, C.F. 2024. Few-shot symbol detection in engineering drawings. Applied artificial intelligence [online], 38(1), article number e2406712. Available from: https://doi.org/10.1080/08839514.2024.2406712

Recently, there has been significant interest in digitizing engineering drawings due to their complexity and practical benefits. Symbol digitization, a critical aspect in this field, is challenging as utilizing Deep Learning-based methods to recogniz... Read More about Few-shot symbol detection in engineering drawings..

Which classifiers are connected to others? An optimal connection framework for multi-layer ensemble systems. (2024)
Journal Article
DANG, T., NGUYEN, T.T., LIEW, A.W.-C., ELYAN, E. and MCCALL, J. 2024. Which classifiers are connected to others? An optimal connection framework for multi-layer ensemble systems. Knowledge-based systems [online], 304, article number 112522. Available from: https://doi.org/10.1016/j.knosys.2024.112522

Ensemble learning is a powerful machine learning strategy that combines multiple models e.g. classifiers to improve predictions beyond what any single model can achieve. Until recently, traditional ensemble methods typically use only one layer of mod... Read More about Which classifiers are connected to others? An optimal connection framework for multi-layer ensemble systems..

A multiclass imbalanced dataset classification of symbols from piping and instrumentation diagrams. (2024)
Presentation / Conference Contribution
JAMIESON, L., MORENO-GARCÍA, C.F. and ELYAN, E. 2024. A multiclass imbalanced dataset classification of symbols from piping and instrumentation diagrams. In Barney Smith, E.H., Liwicki, M. and Peng, L. (eds.) Proceedings of the 18th International conference on Document analysis and recognition 2024 (ICDAR 2024), 30 August - 4 September 2024, Athens, Greece. Lecture notes in computer science, 14804. Cham: Springer [online], part 1, pages 3-16. Available from: https://doi.org/10.1007/978-3-031-70533-5_1

Engineering diagrams provide rich source of information and are widely used across different industries. Recent years have seen growing research interest in developing solutions for processing and analysing these diagrams using wide range of image-pr... Read More about A multiclass imbalanced dataset classification of symbols from piping and instrumentation diagrams..

Towards automated remote inspection of anomalies in offshore components. (2024)
Thesis
TORAL QUIJAS, L.A. 2024. Towards automated remote inspection of anomalies in offshore components. Robert Gordon University, MRes thesis. Hosted on OpenAIR [online]. Available from: https://doi.org/10.48526/rgu-wt-2801306

This dissertation marks a significant advancement in offshore structural inspections, focusing on the development, integration and evaluation of advanced deep-learning models. The research encompasses: a thorough literature review, identifying innova... Read More about Towards automated remote inspection of anomalies in offshore components..

Towards fully automated processing and analysis of construction diagrams: AI-powered symbol detection. (2024)
Journal Article
JAMIESON, L., MORENO-GARCIA, C.F. and ELYAN, E. 2025. Towards fully automated processing and analysis of construction diagrams: AI-powered symbol detection. International journal on document analysis and recognition [online], 28, pages 71-84. Available from: https://doi.org/10.1007/s10032-024-00492-9

Construction drawings are frequently stored in undigitised formats and consequently, their analysis requires substantial manual effort. This is true for many crucial tasks, including material takeoff where the purpose is to obtain a list of the equip... Read More about Towards fully automated processing and analysis of construction diagrams: AI-powered symbol detection..

A multimodel-based screening framework for C-19 using deep learning-inspired data fusion. (2024)
Journal Article
SHANKAR, A., RIZWAN, P., MEKALA, M.S., ELYAN, E., GANDOMI, A.H., MAPLE, C. and RODRIGUES, J.J.P.C. 2024. A multimodel-based screening framework for C-19 using deep learning-inspired data fusion. IEEE journal of biomedical and health informatics [online], Early Access. Available from: https://doi.org/10.1109/JBHI.2024.3400878

In recent times, there has been a notable rise in the utilization of Internet of Medical Things (IoMT) frameworks particularly those based on edge computing, to enhance remote monitoring in healthcare applications. Most existing models in this field... Read More about A multimodel-based screening framework for C-19 using deep learning-inspired data fusion..

Artificial intelligence-based conversational agents used for sustainable fashion: systematic literature review. (2024)
Journal Article
HERNANDEZ MANZO, D.S., JIANG, Y., ELYAN, E. and ISAACS, J. 2025. Artificial intelligence-based conversational agents used for sustainable fashion: systematic literature review. International journal of human-computer interaction [online], 41(8), pages 4640-4652. Available from: https://doi.org/10.1080/10447318.2024.2352920

In the past five years, the textile industry has undergone significant transformations in response to evolving fashion trends and increased consumer garment turnover. To address the environmental impacts of fast fashion, the industry is embracing art... Read More about Artificial intelligence-based conversational agents used for sustainable fashion: systematic literature review..

Segmentation framework for heat loss identification in thermal images: empowering Scottish retrofitting and thermographic survey companies. (2024)
Presentation / Conference Contribution
HASAN, M.J., ELYAN, E., YAN, Y., REN, J. and SARKER, M.M.K. 2024. Segmentation framework for heat loss identification in thermal images: empowering Scottish retrofitting and thermographic survey companies. In Ren, J., Hussain, A., Liao, I.Y. et al. (eds.) Advances in brain inspired cognitive systems: proceedings of the 13th International conference on Brain-inspired cognitive systems 2023 (BICS 2023), 5-6 August 2023, Kuala Lumpur, Malaysia. Lecture notes in computer sciences, 14374. Cham: Springer [online], pages 220-228. Available from: https://doi.org/10.1007/978-981-97-1417-9_21

Retrofitting and thermographic survey (TS) companies in Scotland collaborate with social housing providers to tackle fuel poverty. They employ ground-level infrared (IR) camera-based-TSs (GIRTSs) for collecting thermal images to identify the heat los... Read More about Segmentation framework for heat loss identification in thermal images: empowering Scottish retrofitting and thermographic survey companies..

A review of deep learning methods for digitisation of complex documents and engineering diagrams. (2024)
Journal Article
JAMIESON, L., MORENO-GARCIA, C.F. and ELYAN, E. 2024. A review of deep learning methods for digitisation of complex documents and engineering diagrams. Artificial intelligence review [online], 57(6), article number 136. Available from: https://doi.org/10.1007/s10462-024-10779-2

This paper presents a review of deep learning on engineering drawings and diagrams. These are typically complex diagrams, that contain a large number of different shapes, such as text annotations, symbols, and connectivity information (largely lines)... Read More about A review of deep learning methods for digitisation of complex documents and engineering diagrams..

DICAM: deep inception and channel-wise attention modules for underwater image enhancement. (2024)
Journal Article
FARHADI TOLIE, H., REN, J. and ELYAN, E. 2024. DICAM: deep inception and channel-wise attention modules for underwater image enhancement. Neurocomputing [online], 584, article number 127585. Available from: https://doi.org/10.1016/j.neucom.2024.127585

In underwater environments, imaging devices suffer from water turbidity, attenuation of lights, scattering, and particles, leading to low quality, poor contrast, and biased color images. This has led to great challenges for underwater condition monit... Read More about DICAM: deep inception and channel-wise attention modules for underwater image enhancement..

Generalisation challenges in deep learning models for medical imagery: insights from external validation of COVID-19 classifiers. (2024)
Journal Article
HAYNES, S.C., JOHNSTON, P. and ELYAN, E. 2024. Generalisation challenges in deep learning models for medical imagery: insights from external validation of COVID-19 classifiers. Multimedia tools and applications [online], 83(31), pages 76753-76772. Available from: https://doi.org/10.1007/s11042-024-18543-y

The generalisability of deep neural network classifiers is emerging as one of the most important challenges of our time. The recent COVID-19 pandemic led to a surge of deep learning publications that proposed novel models for the detection of COVID-1... Read More about Generalisation challenges in deep learning models for medical imagery: insights from external validation of COVID-19 classifiers..

All Outputs (104)