AIDA Health

We are dedicated to shaping the future of health technology by harnessing the power of AI to revolutionise medical diagnostics, treatment planning, and patient monitoring, ultimately improving healthcare in the community. With a multidisciplinary team of experts in machine learning, biomedical engineering, and data science, we are at the forefront of developing innovative technologies that tackle some of the most pressing challenges in modern medicine and healthcare.

At AIDA-Health, we focus on combining hardware and software solutions that integrate seamlessly with existing healthcare systems for “doctorless hospitals”, enhancing their efficiency and accuracy. Our research spans various domains, from wearable devices for brain monitoring and seizure detection to in-ear ECG sensors that identify chronic cardiac conditions and specialised hardware for eHealth. Through collaborations with hospitals, industry partners, and academic institutions worldwide, we are committed to transforming our groundbreaking research into practical applications that provide continuous, unobtrusive solutions for both patients and healthcare providers.

We strongly believe in the power of democratising healthcare through open-source AI, making our technologies more accessible to a wider audience. Additionally, our commitment to ethical AI development ensures that all our solutions prioritise patient privacy and safety. 

Topics

  • Hearables: In-ear Monitoring of Neural Function and Vital Signs

    Developing innovative in-ear hardware with specialized sensors to continuously monitor brain activity and physiological parameters such as heart rate and oxygen levels. These easy-to-wear devices provide real-time data and alerts, enhancing the ability to detect and respond to changes in a patient's neurological and overall health status.

  • Safeguarded Explanable AI for Health Analytics

    Combining deep learning approaches with algorithmic prompting of finely-tuned Large Language Models (LLMs) to create safe and efficient diagnostic tools. This involves managing uncertainties arising from environmental noise, physiological variability, and limitations in medical data, ensuring robust data representations, and establishing interpretability in multi-step reasoning by LLMs.

  • Generative AI for Biosignal Processing

    Utilizing generative AI to process complex biosignals, accounting for non-stationary data, multivariate sources, and real-time operational needs. This approach is crucial for extracting meaningful insights from extensively noisy health data via artifact classification, removal, etc. These techniques improve the understanding and management of various health conditions, leading to more accurate and reliable health assessments.

  • Bioboard: Universal Platform for Recording Biosignals

    Leveraging AI algorithms and specialised hardware to record medical data for early detection and diagnosis of diseases, significantly enhancing the accuracy and speed of clinical decision-making. These systems streamline diagnostic processes, providing healthcare professionals with 24/7 real-time insights, thereby improving patient treatment and outcomes.

  • eHealth and Digital Health

    Advancing digital health technologies through the application of AI techniques such as adversarial training, reinforcement learning, and instruction tuning. These methods enhance the reliability and ethical alignment of AI models, ensuring accurate and trustworthy digital health solutions. This includes improving online diagnostics, enabling more precise and real-time health assessments, and providing patients and healthcare providers with actionable insights for better health management.

  • Personalised Medicine Design and Delivery

    Employing AI and big data analytics to design and deliver personalized medicine. By uncovering patterns in health data, this approach enables predictive analytics and personalized treatment plans, enhancing patient outcomes and preventative care strategies. Additionally, it facilitates the identification of novel biomarkers for individuals, contributing to the development of new diagnostics and therapies tailored to individual patient profiles.

Selected Publications from AIDA-Health

Other Publications from AIDA-Health

Artificial intelligence-enhanced electrocardiography derived body mass index as a predictor of future cardiometabolic disease

An actionable, explainable, and biologically plausible AI-ECG risk estimation platform for diabetes mellitus

Artificial intelligence enabled electrocardiogram for mortality and cardiovascular risk estimation: An actionable, explainable and biologically plausible platform

Clinical translation of machine learning algorithms for seizure detection in scalp electroencephalography: a systematic review

From Scalp to Ear-EEG: A Generalizable Transfer Learning Model for Automatic Sleep Scoring in Older People

Hearables: In-ear Multimodal Data Fusion for Robust Heart Rate Estimation

Hearables: feasibility of recording cardiac rhythms from single in-ear locations

Machine learning-based classification of arterial spectral waveforms for the diagnosis of peripheral artery disease in the context of diabetes: A proof-of-concept study