Digital twins support cross-modal and cross-centric classification of mild cognitive impairment

Neural recordings capture crucial pathophysiological processes along the dementia continuum. However, cross-center variability in recording techniques and paradigms limit their generalizability and diagnostic power, preventing clinical use. We here propose a computational approach enabling cross-center classification even in the presence of completely different clinical pipelines. We leveraged a digital twin model to derive digital biomarkers linking neurodegeneration mechanisms to alterations in neural activity across multiple recording modalities. We tested the generalizability of digital biomarkers through cross-center classification of Mild Cognitive Impairment (MCI) and healthy subjects in two independent clinics. The two datasets presented different recording techniques (EEG and MEG), preprocessing modalities, recruitment criteria and diagnostic guidelines. Digital biomarkers derived from one clinic were tested for classifying patients in the other clinic and vice versa employing a transfer learning approach. Digital biomarkers outperform standard biomarkers in the MCI vs healthy classification in both separate datasets (83% vs 58% for EEG dataset and 75% vs 68% for MEG dataset). Moreover, they achieve accurate and consistent cross-center classification (77–78% accuracy), while standard biomarkers perform poorly in the generalization attempt (56–65%). Additionally, digital biomarkers reliably predict global cognitive status across clinics across both datasets ( p < 0.01), while standard biomarkers present no correlation. Digital biomarkers generalize across recording techniques and datasets, enabling a cross-modal and cross-center classification of a patient’s condition. These biomarkers offer a robust measure of patient-specific neurodegeneration, mapping neural recordings anomalies into a common framework of underlying structural alterations. The vast differences between the two datasets support the applicability of this approach also in the presence of high inter-center variability. Amato et al. evaluate whether brain digital twins can be utilized to derive robust dementia biomarkers. Biomarkers derived from digital twin outperform standard MEG and EEG metrics in diagnosing pathological cognitive decline across different clinical centers. People with pathological cognitive decline lose memory, reasoning and communication ability, developing diseases such as dementia. Non-invasive methods that record activity in the brain have been investigated to see whether they can be used to identify people with pathological cognitive decline. Ideally these methods would require minimal equipment and reduce patient discomfort, but their results often vary greatly between clinics. We used patient-specific computational brain models, called digital brain twins, to find possible digital markers of pathological cognitive decline. We evaluated whether these markers could be used to diagnose pathological clinical cognitive decline across two groups of people in different clinics. Our results demonstrate that digital markers generalize well across different clinical settings and recording methods. They could provide better indicators of pathological cognitive decline than currently used alternatives and so improve identification of people with pathological cognitive decline.