This research introduces a multimodal learning framework combining wearable sensor streams and electronic medical records to predict cardiovascular disease risk. It enhances model robustness by aligning physiological trends with clinical profiles, supporting continuous monitoring and proactive care through cloud-integrated predictive systems.

Lawrence Patrick Whitmore, Emily Frances Hawthorne, Theodore Marcus Burnham, Hannah Louise Sutherland, Benjamin Oscar Chadwick

This paper presents a blockchain-integrated cloud solution for managing biomedical imaging records in telemedicine. The proposed system enhances security and transparency, ensuring traceability and integrity of imaging data while facilitating access control and auditing across multiple healthcare service providers.

Charles Anthony Redford, Georgia Isabelle Whitaker, Felix Jonathan Morland, Victoria Eleanor Chambers, Zachary Daniel Prescott

This paper proposes a reinforcement learning approach to resource management in genomic data analysis platforms. Coupled with container orchestration, the model ensures efficient computing resource distribution, minimizing latency while processing large-scale sequencing data in biomedical research environments hosted on the cloud.

Harvey Douglas Wainwright, Sophie Mae Underwood, Lucas Franklin Medford, Isabelle Grace Linton, Owen Gregory Fairchild

This study designs a cloud-hosted time-series model for early detection of post-surgical infections. Utilizing recurrent neural networks on vital sign patterns, the system enables clinicians to monitor and intervene earlier, potentially reducing complications, hospital stays, and healthcare costs in recovery pathways.

Daniel Thomas Wilkes, Alice Margaret Liddell, William Henry Sloane, Charlotte Anne Forsythe, Matthew Edward Crowley

This paper introduces a scalable data lake architecture to unify biomedical research data. By integrating multi-modal health records across institutions, the platform enhances interoperability, supports collaborative analysis, and accelerates research insights with cloud-native tools and schema-flexible data ingestion techniques.

Leonard James Brinkley, Phoebe Harriet Carleton, Sebastian Oliver Tisdale, Madeleine Louise Kemp, Joseph Benjamin Hadley

This research proposes federated learning models for privacy-focused clinical decision support. Operating in decentralized cloud environments, the system enables collaborative training without sharing raw data, maintaining patient confidentiality while enhancing prediction accuracy for diverse medical applications across institutions and geographic regions.

Maxwell George Trenholm, Olivia Beatrice Langston, Hugo Alexander Donnelly, Amelia Rose Bannister, George Francis Cattermole

This paper outlines a system for real-time EEG monitoring using wearable devices connected to cloud systems via edge nodes. The architecture supports latency-sensitive data transmission and storage, allowing continuous assessment of neurological health while optimizing network traffic and preserving user privacy.

Isaac Malcolm Rutherford, Eleanor Grace Fenwick, Harvey Nicholas Millwood, Georgia Faith Carling, Thomas Nathaniel Warburton

This study develops cloud-hosted AI models for detecting diabetic retinopathy from high-resolution fundus images. Utilizing convolutional neural networks, the system enhances diagnostic accuracy and provides scalable, remote screening capabilities, improving early intervention and management of diabetic eye conditions worldwide.

Frederick Ellis Barlow, Matilda Iris Northcott, Samuel Lewis Hollingsworth, Rebecca Clare Whittemore, Edward James Godfrey