MemSpark: An Artificially Intelligent Virtual Reality System for Non-Intrusive Cognition Therapy and Evaluation of Dementia

Dementia will become an epidemic with cases projected to grow from 55 million to 153 million by 2050. Despite such severity, mainstream therapeutic options fail to affect progression as most are symptomatic, invasive, or ineffective. This research explored a novel virtual reality based therapy that can exercise all cognitive functions to slow dementia progression. Improvements in cognition were measured using three aggregate behavioral biomarkers: recall, reasoning, and executive function. An artificial neural network was developed to compute dementia profiles aligned to the ADAS-Cog test. Trials conducted across 14 people with dementia (PwD) over six months indicated 60% slower rate of deterioration among the experimental group. The Prolonged Life Expectancy metric analysis suggested that this therapy could double the time PwDs have before reaching significant cognitive impairment. This cognition therapy has the potential to replace mainstream therapies such as tDCS and CCT and eliminate dependency on dementia experts to enable in-home care.