Search papers, labs, and topics across Lattice.
This paper introduces Delta2Gamma, a self-supervised framework that enhances EEG representation learning for Alzheimer's disease detection by contrasting augmented views of EEG signals decomposed into five canonical neural rhythms. By employing adaptive temperature scaling during contrastive training, the method effectively balances the varying signal statistics across different frequency bands. The approach achieves a remarkable 92.4% accuracy in distinguishing Alzheimer's patients from cognitively normal controls, outperforming both supervised methods and existing EEG-specific techniques.
Delta2Gamma achieves a striking 92.4% accuracy in Alzheimer's detection using low-cost EEG, surpassing traditional supervised methods and dedicated EEG techniques.
Low-cost, scalable screening for dementia remains an open problem. Imaging-based diagnosis is costly and hard to deploy widely. Electroencephalography (EEG) is portable and inexpensive, but its recordings are noisy, vary widely across subjects, and carry few clinical labels. We tackle this with Delta2Gamma, a self-supervised framework that learns EEG representations from unlabeled data by contrasting augmented views of each signal. Rather than treat EEG as a single stream, Delta2Gamma decomposes every recording into the five canonical neural rhythms (delta, theta, alpha, beta, gamma). Each band gets its own encoder and projection head. Each also gets a temperature that is predicted adaptively during contrastive training, so bands with different signal statistics are balanced automatically. On the ADFTD cohort under a strict leave-one-subject-out protocol, Delta2Gamma separates Alzheimer's disease from cognitively normal controls with 92.4\% accuracy. This exceeds both supervised backbones and recent dedicated EEG methods.
