Artificial Intelligence in Epilepsy

Home > Artificial Intelligence in Epilepsy

Artificial intelligence or AI is the ability of machines or technology to mimic human intelligence, especially computer systems. So, when your smartphone auto-corrects or an online store sends you advertisements for a new product it thinks you’ll like, you can say thank you to AI.

AI is being used in a wide range of industries and applications, including healthcare, finance, and transportation. Examples in healthcare include; AI is used to analyse medical data and images, and make diagnoses, in finance; AI is being used for fraud detection and financial forecasting, and in transportation; AI is being used in development of self-driving cars and matters such as improving traffic safety.

In healthcare, AI is used more and more to aid in the diagnosis, treatment, and management of disease, including epilepsy. It’s important to note that while AI shows promise in epilepsy management, it is still very much an evolving field, and further research and validation are needed to ensure its accuracy and clinical effectiveness.

AI Applications in healthcare and epilepsy

The potential for using AI in healthcare is limitless and promises a future filled with innovations, improved health outcomes and better experiences for patients. Some AI applications used in epilepsy include:

EEG Interpretation

AI techniques can assist in interpretation of EEG recordings by analysing large chunks of EEG recording in a much shorter space of time (useful in overnight or several days of recordings). This will then be further interpreted by the neurologist. AI can help clinicians with diagnosis by identifying abnormalities, locating the seizures and seizure focus (the area of the brain where seizures originate), distinguishing epileptic activity from normal brain activity, and provide timely and appropriate management or intervention.

Seizure Prediction

AI algorithms can analyse electroencephalogram (EEG) data over a lengthy period to identify patterns and predict when seizures are more likely to occur. By detecting pre-seizure patterns, AI systems are able provide advance warnings when a seizure is more likely occur, allowing people with epilepsy to take preventive measures. This is also called seizure forecasting, and is available through the Seer App

Seizure Detection

AI algorithms can be trained to automatically detect seizure activity in EEG recordings. This can aid in the early identification of seizures, allowing for timely interventions.

Diagnostic Images

AI offers a promising approach to measure abnormalities in neuroimaging data such as brain imaging. This should improve the detection of brain abnormalities and seizure localisation, an important tool in diagnosis and management of epilepsy.

Treatment

Neuromodulation therapy using various brain stimulation techniques, including vagal nerve stimulation (VNS) and deep brain stimulation (DBS), are sometimes used in the management of medication resistant epilepsy. These therapies modulate nerve and brain function by delivering small electrical stimulation or pharmaceutical agents directly to a target area.

Medication

The potential applications of AI for medication management are vast, providing many benefits for both pharmacists and patients. Examples of AI benefits with medication include improved patient safety and prevention of medication errors with dispensing, or its use as a medication reminder for patients to remember to take medication.

Risk Assessment

AI models can assess the risk of seizure occurrence based on various aspects, such as taking medication, sleep patterns, stress levels, and environmental triggers. This information can assist in developing personalised epilepsy management plans and lifestyle modifications.

Treatment Decision Support

AI systems can aid clinicians in making treatment decisions by analysing a patient’s medical history, EEG data, genetic information, and response to previous treatments. This can help determine the most effective treatment options for individual patients. Also known as clinical decision support systems, CDSS, is where you have a shared interface where you can focus on specific data such as when evaluating epilepsy surgical treatment, you can integrate all the data: MRI, PET scans, SPECT, EEG, relatively quickly and easily enough to be usable and make a clinical decision (Bosselmann 2023).

Wearable Devices

AI is used in wearable devices such as smartwatches or seizure detection devices, to monitor physical signals (such as seizure-like movements, heart rate, blood oxygen levels, falls) and detect seizures in real-time. These devices can alert caregivers to a seizure, which reduces risks to the person with epilepsy. (An et al 2020, Ong et al 2022)

The following are a few areas in epilepsy where AI has been found to be helpful:

  • Automatic seizure detection and prediction
  • Diagnosis and classification of seizures and epilepsies
  • Understanding the cause of the epilepsy
  • Optimising the medical and surgical management
  • Development of wearable electronic devices for PWE (Nair et al 2021)

A recent case study showed that detection of epileptic seizures by deep learning in low-power neural implants was comparable to state-of-the-art algorithms that run in high-performance computers. The researcher, Liu, says that his team’s technology can be used in a broad range of clinical applications beyond epilepsy, noting that up to one billion people worldwide suffer from various brain disorders. (University of Toronto 2022)

AI Challenges

AI has the potential to bring many benefits, but we can’t ignore that there are, and will be, challenges. Issues such as cybersecurity, data privacy, and the need for robust regulations to ensure the safe and ethical deployment of AI systems need to be addressed and managed.

However, the benefits of AI in healthcare far outweigh the potential limitations and challenges, paving the way for a brighter future.

Conclusion

AI is already playing a significant role in medical diagnosis, treatment and prognosis. It is going to play a far bigger role in the future. It has the potential to shape diagnosis, prediction, and management of neurological diseases such as epilepsy, and to improve patient engagement and health education and promotion. AI is not intended to replace medical opinion but to transform healthcare into a system where technology and humans can work together to provide an accurate, timely diagnosis, and management of the health condition. (Babel etal 2021)

Definitions:

Algorithm: a set of instructions for solving a problem or completing a task.

Neuroimaging: the process of producing images of the structure or function of the brain or other part of the nervous system by techniques such as magnetic resonance imaging (MRI) or computerised tomography (CT).

References
Artificial Intelligence 101: Understanding the Basics of AI https://www.linkedin.com/pulse/artificial-intelligence-101-understanding-basics-ai/ Accessed 22 Aug 2023
An, S., Kang, C., & Lee, H. W. (2020). Artificial Intelligence and Computational Approaches for Epilepsy. Journal of epilepsy research, 10(1), 8–17.
Babel, A., Taneja, R., Mondello Malvestiti, F., Monaco, A., & Donde, S. (2021). Artificial Intelligence Solutions to Increase Medication Adherence in Patients With Non-communicable Diseases. Frontiers in digital health, 3, 669869.
Bosselmann, C. (2023) Artificial intelligence and epilepsy: an interview on Sharpwaves. Epigraph Vol. 25 Issue 2. https://www.ilae.org/journals/epigraph/epigraph-vol-25-issue-2-spring-2023/artificial-intelligence-and-epilepsy-dr-christian-bosselmann Accessed 22 Aug 2023
Choi, K. S., & Sunwoo, L. (2022). Artificial intelligence in neuroimaging: clinical applications. Investigative Magnetic Resonance Imaging, 26(1), 1-9.
Gomez-Quintana, S., O’Shea, A., Factor, A., Popovici, E., & Temko, A. (2022) A method for AI assisted human interpretation of neonatal EEG. Scientific Reports. 12, 10932.
Kaur T, Diwakar A, Kirandeep, Mirpuri P, Tripathi M, Chandra P S, Gandhi TK. Artificial Intelligence in Epilepsy. Neurol India [online] 2021;69:560-6.
Nair, P. P., Aghoram, R., & Khilari, M. L. (2021). Applications of artificial intelligence in epilepsy. International Journal of Advanced Medical and Health Research, 8(2), 41-48.
Ong, J. S., Wong, S. N., Arulsamy, A., Watterson, J. L., & Shaikh, M. F. (2022). Medical Technology: A Systematic Review on Medical Devices Utilized for Epilepsy Prediction and Management. Current neuropharmacology, 20(5), 950–964. https://doi.org/10.2174/1570159X19666211108153001
Osama Khan, Mohd Parvez, Pratibha Kumari, Samia Parvez, Shadab Ahmad. (2023) The future of pharmacy: How AI is revolutionizing the industry, Intelligent Pharmacy, Volume 1, Issue 1, Pages 32-40, ISSN 2949-866X, https://doi.org/10.1016/j.ipha.2023.04.008
Forecasting seizure cycles https://pursuit.unimelb.edu.au/articles/forecasting-the-cycle-of-epileptic-seizures
University of Toronto News. Oct 17 2022. Researcher combines AI and microelectronics to create neural implants that fight brain disorders. https://www.utoronto.ca/news/researcher-combines-ai-and-microelectronics-create-neural-implants-fight-brain-disorders  Accessed 22 Aug 2023

Get In Touch With Us

    We acknowledge the Traditional Owners of the land where we work and live. We pay our respects to Elders past, present and emerging.

    We celebrate the stories, culture and traditions of Aboriginal and Torres Strait Islander Peoples of all communities who also work and live on this land.

    We are committed to providing a child-safe environment, prioritising the well-being and protection of every child in our care.

    Important Message: This information is not intended as medical advice. Do not make changes to treatment or medications without first discussing with your treating clinician.