New AI reporting tool to improve epilepsy diagnosis and management

A low-cost reporting tool is to be developed using AI to improve epilepsy diagnosis and management, thanks to Monash Institute of Medical Engineering (MIME) Seed Funding.Epilepsy affects 80 million people worldwide with more than 150,000 Australians living with the condition. The annual cost associated with epilepsy is over 12 billion dollars.Diagnosis of epilepsy relies heavily on the analysis of brainwave activity (electroencephalography, or EEG). This process is currently done manually. AI offers the potential to automate the analysis and reporting of brainwave activity, to make it not only faster and more accurate but also able to standardize current EEG reports.This project will develop an EMS, an EEG Management System. It will use cutting-edge AI technology to create reliable biomarker detection and sub-categorization whilst automating and digitalizing EEG reporting. It will use online technology to ensure a low-cost, easily accessible, highly secured and user-friendly option for clinicians.Lead researchers, Dr Duong Nhu, a researcher with an interest in epilepsy from the Faculty of Information Technology at Monash University; and Dr Hugh Simpson, a neurologist with an interest in AI from Alfred Health believe that using AI in epilepsy diagnosis will result in better outcomes for both patients and clinicians.“EEG is critical to diagnosis and monitoring in epilepsy, which are key requirements for management of epilepsy. However, the current EEG analysis process is time-consuming, taking up to several hours for neurologists to analyze and complete the reports. AI can dramatically improve the speed of review whilst dealing with copious amounts of data, ultimately improving patient care for people with epilepsy,” said Dr Nhu.A further benefit of this project’s technology is accessibility.“The World Health Organization's Intersectoral Global Action Plan for epilepsy and other neurological disorders aims to increase access to epilepsy services and diagnostics throughout the world. Automated analysis, which the EEG Management System offers, greatly reduces the amount of clinician review required. This is appealing for large centers, to reduce the workload on neurologists, while the automated reports generated may be particularly appealing for centers that do not have access to neurologists; increasing accessibility to diagnostic testing for people regardless of their location,” said Dr Simpson.The impact of the tool will be significant for clinicians, hospitals, and patients.“The idea is to make automated EEG analysis a more clinically available tool to help patients get better and faster diagnosis and make EEG reporting more structured and more accessible. This doesn’t exist at the moment,” said Dr Nhu.“Currently, we are working with clinicians at Alfred Health and have estimated that with our product between $250-$500K on diagnostic costs could be saved. With very few commercially available products in this space, we are excited to develop a unique low-cost alternative to make a real impact in the lives of patients with epilepsy and in the practice of their treating clinicians,” said Dr Simpson.The project was awarded $40,000 in MIME Seed Funding in late 2023.“Receiving MIME Seed Funding will help to progress this project and cover the costs associated with data collection and annotation, algorithm development, and computing resources,” said Dr Nhu.The project is currently in the early stages with the view to introduce the project to Alfred Health and The Royal Melbourne Hospital by the end of 2025. Watch this space for further developments.