{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/5a413159256a41c99f9a49b9e0b514ad\" frameborder=\"0\" width=\"1668\" height=\"1251\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1251,"width":1668,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1251,"thumbnail_width":1668,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/5a413159256a41c99f9a49b9e0b514ad-9fbb9ca169b9568e.gif","duration":298.816,"title":"Paperclip Orchestrator for Cognitive Impairment Evidence Digest","description":"This Loom presents a Paperclip-based evidence digest to identify wearable signals correlating with cognitive impairment and select the most accurate sensory hardware. The creator copies a question into Cloud Code, runs a pipeline that takes about half an hour, and produces a scaffolded evidence digest organized by sub-questions. Across studies, the strongest and most consistent signals are gait and mobility metrics and circadian rest activity metrics, with delirium showing a partial carve-out. The synthesis also notes that the literature has not formally connected hardware measurement accuracy to downstream cognitive detection performance, and mentions a bug where DOIs can be hallucinated due to non-deterministic processing."}