Researchers at Florida Atlantic University have developed a novel deep learning model that distinguishes between Alzheimer’s disease (AD) and frontotemporal dementia (FTD) using advanced analysis of electroencephalography (EEG) brainwave data. This innovative approach enhances diagnostic accuracy and offers significant implications for patient care and treatment strategies.
Dementia encompasses various disorders that progressively impair cognitive functions, including memory and daily activities. As the most prevalent form of dementia, Alzheimer’s disease is projected to affect approximately 7.2 million Americans aged 65 and older by 2025. In contrast, frontotemporal dementia, while less common, represents the second leading cause of early-onset dementia, affecting individuals typically between their 40s and 60s. Accurate differentiation between these two diseases is crucial, as they exhibit overlapping symptoms yet require distinct treatment approaches.
Advancing Diagnostic Techniques
Traditional diagnostic methods such as MRI and PET scans can be effective but are often costly, time-consuming, and require specialized equipment. EEG presents a portable, noninvasive alternative that measures brain activity through various frequency bands. Despite its advantages, EEG signals can be noisy and variable, complicating analysis and making it challenging to differentiate between dementia types.
To overcome these obstacles, the research team has created a deep learning model that analyzes both frequency and time-based patterns in brain activity. The findings, published in the journal Biomedical Signal Processing and Control, reveal that slow delta brain waves serve as significant biomarkers for both AD and FTD, particularly in the frontal and central regions of the brain. Notably, the model achieved over 90% accuracy in distinguishing between dementia patients and cognitively healthy individuals.
The researchers found that brain activity in Alzheimer’s patients exhibited widespread disruptions, impacting multiple regions and frequency bands, including beta waves. This extensive damage may explain why Alzheimer’s is typically easier to diagnose compared to FTD, which primarily affects specific regions associated with behavior and language.
Improving Diagnostic Precision
By employing feature selection techniques, the research team improved the model’s specificity for identifying healthy individuals from 26% to 65%. Their two-stage design, first identifying healthy participants and subsequently differentiating AD from FTD, achieved an impressive 84% accuracy rate. This performance ranks the model among the most effective EEG-based diagnostic methods available.
The model integrates convolutional neural networks with attention-based long short-term memory networks (LSTMs), enabling it to assess both the type and severity of dementia through EEG data. The use of Grad-CAM technology also allows clinicians to visualize which brain signals influence the model’s predictions, providing critical insights into the diagnostic process.
Tuan Vo, the first author of the study and a doctoral student at the FAU Department of Electrical Engineering and Computer Science, emphasized the study’s innovative approach: “What makes our study novel is how we used deep learning to extract both spatial and temporal information from EEG signals. By doing this, we can detect subtle brainwave patterns linked to Alzheimer’s and frontotemporal dementia that would otherwise go unnoticed.”
The research highlights that Alzheimer’s disease generally leads to more severe cognitive decline, affecting a broader range of brain areas compared to FTD, which has a more localized impact. These insights align with previous findings from neuroimaging studies but enhance understanding by demonstrating how these patterns manifest in EEG data, making it a valuable tool for diagnostics.
Stella Batalama, Ph.D., dean of the College of Engineering and Computer Science, remarked on the potential of this research, stating, “This work demonstrates how merging engineering, AI, and neuroscience can transform how we confront major health challenges. With millions affected by Alzheimer’s and frontotemporal dementia, breakthroughs like this open the door to earlier detection, more personalized care, and interventions that can truly improve lives.”
The study’s co-authors include Ali K. Ibrahim, Ph.D., an assistant professor, and Chiron Bang, a doctoral student, both affiliated with the FAU Department of Electrical Engineering and Computer Science. The research showcases the powerful role of technology in enhancing dementia diagnosis, offering hope for improved patient outcomes and management of these complex disorders.
For further information, refer to the study by Tuan Vo et al., titled “Extraction and interpretation of EEG features for diagnosis and severity prediction of Alzheimer’s Disease and Frontotemporal dementia using deep learning,” published in Biomedical Signal Processing and Control in 2026.