3-minute EEG Test Reliably Detects Memory Impairment, Enabling Earlier Alzheimer's Diagnosis

Breakthrough research from the University of Bath and the University of Bristol that demonstrates the real-world efficacy of a passive EEG test in identifying memory decline years before a clinical diagnosis, is being developed as a screening tool by Cumulus Neuroscience

BELFAST, Northern Ireland, Sept. 2, 2025 /PRNewswire/ -- Cumulus Neuroscience, a global digital health company focused on advancing neuroscience clinical trials and patient care through improved data and AI, announced new research published in the journal Brain Communications, led by Dr. George Stothart, a cognitive neuroscientist based in the Department of Psychology at the University of Bath and Principal Scientist at Cumulus Neuroscience. Dr. Stothart has successfully trialled the Fastball EEG technology, the core component of Cumulus's AccelADx^® screening tool in development, for the first time outside of research labs.

The study found that the three-minute, non-invasive Fastball EEG test can reliably identify early memory impairment in individuals with Mild Cognitive Impairment (MCI), a condition often preceding Alzheimer's disease dementia. This ground-breaking finding builds on previous research and, for the first time, validates the technology's ability to deliver reliable results in a real-world setting, with clinical staff visiting patients in their homes. Cumulus is developing AccelADx, a scalable version of Fastball EEG. AccelADx is currently deployed in two large grant-funded studies – one that is enrolling 1000 participants in the UK, and one that is enrolling an ethnically diverse population of 1,000 participants in the United States, Canada, and Europe. 

The ability to test for early signs of Alzheimer's outside of a research or hospital environment is a critical step forward, as it paves the way for wider screening and earlier diagnosis in any public-health setting. This is particularly important given the recent breakthroughs in Alzheimer's treatments including donanemab and lecanemab, which are most effective when administered in the earliest stages of the disease. Despite this, a significant number of people with dementia worldwide remain undiagnosed, delaying access to crucial treatments and support.

"We are missing the first 10 to 20 years of cognitive decline in Alzheimer's with current diagnostic tools. This research offers a way to change that," said Dr. George Stothart, who is leading the development of AccelADx. "By detecting memory changes far earlier and more objectively, using a quick and passive test, we can empower healthcare providers to intervene at the most critical time."

AccelADx is a proprietary screening tool built on Cumulus Neuroscience's biopharma-proven NeuLogiq® Platform. It combines an easy-to-use EEG headset with a tablet-based assessment that does not require participants to follow complex instructions or recall information. This passive approach makes the test more objective and accessible, as performance is not impacted by education level or language skills. The test monitors the brain's automatic responses to images, with AI-powered analytics quickly identifying subtle, early changes in brainwave activity.

"This paper is an important addition to the evidence base – not only can Fastball EEG detect some of the earliest decline that leads to dementia, but it can also detect memory impairment which is a key symptom of Alzheimer's Disease," said Brian Murphy, PhD, Co-founder and Chief Scientific Officer at Cumulus Neuroscience. "By removing barriers to diagnosis for all patients regardless of ethnicity or education level, AccelADx has the potential to accelerate enrollment in clinical trials, bringing much needed treatments to patients faster. We're looking forward to learning more about the real-world utility of AccelADx through our two grant-funded studies which will continue primary data collection into 2027, with interim results expected in 2026."  

About the paper
The study, titled "A passive and objective measure of recognition memory in mild cognitive impairment using Fastball memory assessment," was authored by George Stothart, Sophie Alderman, Oliver Hermann (University of Bath); Sam Creavin, and Elizabeth Coulthard (University of Bristol). The research was funded by the Academy of Medical Sciences and supported by the dementia research charity BRACE.

About Cumulus Neuroscience
With a mission to generate the data and insights required to accelerate diagnosis and management of central nervous system (CNS) disorders for millions of patients and caregivers around the world, Cumulus Neuroscience is advancing NeuLogiq®, an AI-based, multi-domain digital biomarker platform to enable better, faster decision making in neurology and neuropsychiatry clinical trials and patient care. Designed for and with 10 of the world's leading pharma companies, the platform enables decentralized trials and is already making a difference in the development of therapies for Alzheimer's Disease, depression and schizophrenia.

Designed to provide an industry-wide standard for real-world measurement of disease progression, Cumulus combines patented technology, in-house expertise and key industry partnerships to capture large amounts of real-world, clinical data repeated over time, across multiple behavioral and physiological domains in the patient's home – all with an EEG headset synced to a novel, tablet-based neuro-assessment platform. Together with machine learning (ML) analytics and the world's largest database of annotated, longitudinal, neurofunctional data, Cumulus simplifies and improves the robustness of neuroscience clinical trials to provide the best and most cost-effective assessment of CNS treatment outcomes.

Logo - https://mma.prnewswire.com/media/2657658/5487619/Cumulus_Logo.jpg