Electronic Emulation of Neurodegenerative Disease Progression on a CMOS-Based Physical Model of Human Connectome

Abstract

The present study depicts a newframework for the emulation of neurodegenerative disease built on a Complementary Metal-Oxide-Semiconductor (CMOS) device-level model of the human connectome, in which nodal dynamics is implemented using analog chaotic oscillators. SPICE simulations performed on this platform, following a multimodal lesioning protocol based on established disease trajectories, allow to replicate disease-specific dynamics and identify functional connectivity biomarkers correlated with Alzheimer’s disease and Frontotemporal dementia staging. Specifically, the progressive alteration of network connectivity mimics anatomical degeneration, characteristic of these neurological conditions, and results in measurable changes in synchronization and dynamics. These include gradual decrease in phase coherence and dynamical complexity, accompanied by a global slowing of oscillatory modes. This is further substantiated by graph-theoretical analyses, showing a decrease in clustering coefficient and global efficiency, as well as an increase in modularity. This reflects a loss of balance between functional segregation and integration, highlighting impaired resilience in the event of structural compromise. Transposing the obtained synchronization patterns to resting-state networks reveals a selective and gradual loss of intra- and inter-network functional connectivity across the disruption stages. The hardware-oriented method used in this work is a unique approach to neurodegenerative disease simulation and represents an effective framework for investigating human connectome vulnerabilities. As an in-silico disease model, it provides a viable and reproducible environment suitable for evaluating future diagnostic biomarkers and therapeutic interventions. Moreover, it is a neuromorphic system that can be integrated with neuroprosthetic devices and closed-loop brain–computer interfaces to support the development of implantable diagnostics or adaptive neuromodulation systems for patient-specific treatment.