Radio Frequency-Based Vascular Dementia Sensing and Imaging System Targeting Smart Glasses

Abstract

Vascular dementia is the second most prevalent type of dementia among the elderly population and is one of the leading causes of mortality. Ischemic stroke and brain atrophy are the predominant pathologies associated with vascular dementia. Early detection and regular monitoring are crucial to prevent the advancement of vascular dementia. Conventional medical imaging is expensive, requires extensive medical supervision and is not easily accessible. This research presents a novel concept of low-cost and noninvasive smart glasses, equipped with miniaturized octagonal monopole-patch antenna (OMPA) sensors and a crescent array sensor for vascular dementia detection. This radio frequency (RF) enabled portable system is capable of accurately identifying and imaging brain infarction, atrophy, and stroke in their early stages. The fabricated device is experimentally verified using multiple artificial stroke and atrophy targets inside a realistic brain phantom. The backscattered RF data is iteratively processed using customized imaging algorithms to achieve improved image quality, noise suppression, and contrast resolution with reduced image artifacts and computational complexity. Based on the iterative refinement, the double-stage minimum variance delay multiply and sum (DS-MV-DMAS) algorithm is proposed for imaging stroke and brain atrophy. The quantitative results indicate that DS-MV-DMAS results in 43% lower level of side lobes and leads to 26%, 28%, and 27% improvement in signal-to-noise ratio (SNR), full width at half-maximum and contrast ratio (CR), respectively, compared to other state-of-the-art (SOTA) imaging algorithms. The promising results demonstrate the feasibility of the prototype system as a cost-effective, portable, and noninvasive alternative for the diagnosis and monitoring of vascular dementia.