Digital Memory Augmentation based on how the brain works (i2i Phase 1)

记忆丧失是由阿尔茨海默病（AD）和通常先于AD的轻度认知损害（MCI）引起的最常见和使人衰弱的认知损害。目前有56.4万加拿大人患有AD或相关痴呆症，每年花费330亿加元（到2040年将增加到2930亿加元）。记忆丧失会削弱一个人的身份感和自主性，导致社会孤立和抑郁的恶性循环，这反过来又加剧了记忆丧失。数字记忆增强（DMA）是一种减轻记忆丧失的方法，其中便携式设备捕获有关日常事件的信息，允许以后回顾和回忆。DMA已被证明能产生记忆益处--甚至比记日记等常见策略更有益处。然而，先前的DMA方法并没有被我们对大脑功能或关键学习原则的理解所告知。本项目旨在优化和现场测试一种新型的DMA平台，以最佳方式减少衰老和痴呆症的记忆障碍。这是第一个基于我们对大脑内部如何“提醒”自己先前事件的理解而设计的DMA工具。它模仿海马体的重要功能-MCI/AD中受损的大脑区域-通过重放和重组个人生活的电影来稳定和检索记忆。拟议的工作将增强重播算法，以提高功效和用户享受。具体来说，我们将使用领域领先的机器视觉工具，以便重播算法将相关事件分组为有意义和连贯的叙述，可以根据需要检索，以揭示与任何给定情况相关的记忆。我们预测，将相关事件分组为与当前情况相关的有意义的叙述，不仅会增强记忆力，还会提高用户满意度和参与度。我们将与主要利益相关者测试我们的增强平台，包括社区居住的老年人以及居住生活设施的居民和护理人员。

Memory loss is the most common and debilitating cognitive impairment resulting from Alzheimer's disease (AD) and from the mild cognitive impairment (MCI) which often precedes AD. Currently 564,000 Canadians live with AD or a related dementia, costing $33 billion annually (rising to $293 billion by 2040). Memory loss diminishes one's sense of identity and autonomy, leading to a vicious cycle of social isolation and depression,which in turn exacerbates memory loss. Digital Memory Augmentation (DMA) is an approach to mitigate memory loss in which portable devices capture information about everyday episodes, allowing for later review and recollection. DMA has been shown to produce memory benefits - even more so than common strategies such as keeping a diary. However, prior DMA approaches have not been informed by our understanding ofbrain function or key learning principles.This project aims to optimize and field test a new kind of DMA platform that optimally reduces memory impairment in aging and dementia. This is the first DMA tool designed based on our understanding of how the brain internally "reminds" itself of prior events. It mimics important functions of the hippocampus - a brain area damaged in MCI/AD - by replaying and reorganizing movies of an individual's life to stabilize and retrieve memories. The proposed work will enhance the replay algorithm to improve efficacy and user enjoyment. Specifically, we will use field-leading machine vision tools so that the replay algorithm groups related events into meaningful and coherent narratives that can retrieved on demand to reveal memories relevant to any given situation. We predict that grouping related events into meaningful narratives that are pertinent to the currentsituation will not only enhance memory benefits, but also lead to increased user satisfaction and engagement. We will test our enhanced platform with key stakeholders, including community-dwelling older adults as well as residents and caregivers at a residential living facility.