发表机构
Lausanne University Hospital; Faculty of Biology and Medicine, University of Lausanne; VNU University of Engineering and Technology; VinUni-Illinois Smart Health Center, VinUniversity; University of Science, VNU-HCM; Hanoi Medical University; National Geriatric Hospital; Military Hospital 175; School of Medicine, University of Medicine and Pharmacy at Ho Chi Minh City; International University, VNU-HCM; Vietnam National University Ho Chi Minh City（洛桑大学医院; 洛桑大学生物与医学学院; 越南国立大学工程技术大学; 文大-伊利诺伊智能健康中心，文大; 胡志明市越南国立大学科学大学; 河内医科大学; 国家老年病医院; 第175军医院; 胡志明市医药大学医学院; 胡志明市越南国立大学国际大学; 胡志明市越南国立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
MemoCare提出交互式多模态移动认知筛查系统，结合语音、空间定向、触摸和绘图任务，通过本地NLP与CNN共识自动评分，测试准确率高且获临床认可。
AI 中文摘要
MemoCare是一种用于自动化多模态认知筛查的交互式移动系统。一个React Native应用程序在完整的英语和越南语工作流程中结合了口语回答、时间和空间定向、触摸屏操作以及视觉构造。语音由Google Speech-to-Text转录，并使用确定性的任务特定自然语言处理规则在本地评分；GPS坐标由MemoCare空间模块在答案匹配之前解析；触摸任务根据交互事件评分；绘图任务使用具有独立视觉解释的三模型卷积神经网络共识。软件测试在语音/语言、空间答案和触摸交互评分方面通过了151/151个预定义案例，而空间回归在48/48个四国坐标解析案例中通过。对于绘图模块，验证选择的ShuffleNetV2 x1.5在锁定的71图像测试集上实现了91.33%的平均平衡准确率和78.87%的精确三标准准确率。四位临床合著者还检查了端到端工作流程，在八个标准上产生了4/5的汇总中位数评分，项目级中位数范围从3到4.5。在MMM上，与会者可以直接尝试缩短的多模态筛查工作流程，并检查自动项目级和总分评分。
英文摘要
MemoCare is an interactive mobile system for automated multimodal cognitive screening. A React Native application combines spoken responses, temporal and spatial orientation, touchscreen actions, and visuoconstruction in complete English and Vietnamese workflows. Speech is transcribed by Google Speech-to-Text and scored locally with deterministic task-specific natural language processing rules; GPS coordinates are resolved by the MemoCare spatial module before answer matching; touch tasks are scored from interaction events; and the drawing task uses a three-model convolutional neural network consensus with separate visual interpretation. Software tests pass 151/151 predefined cases across speech/language, spatial-answer, and touch-interaction scoring, while spatial regression passes 48/48 four-country coordinate-resolution cases. For the drawing module, validation-selected ShuffleNetV2 x1.5 achieved 91.33% mean balanced accuracy and 78.87% exact three-criterion accuracy on a locked 71-image test set. Four clinician co-authors additionally inspected the end-to-end workflow, yielding a pooled median rating of 4/5 across eight criteria, with item-level medians ranging from 3 to 4.5. At MMM, attendees can directly try a shortened multimodal screening workflow and inspect automatic item-level and total scoring.
Comments8 pages, 1 figure, 1 table. Demo paper submitted to the MMM 2027 Demo Track
