ContextDelirium is a significant concern in end-of-life care. Continuous monitoring of agitation levels using objective methods may have advantages over existing measurement scales. ObjectivesTo examine whether an objective measure of activity scores measured using a sheet-type non-wearable sensor (Nemuri SCAN [NSCAN]) was correlated with agitation levels measured using the modified Richmond Agitation-Sedation Scale (RASS) in terminally ill patients with cancer. MethodsWe conducted a single-center, prospective, observational study in a palliative care unit using the NSCAN to measure activity scores and the RASS to assess agitation levels. RASS scores were prospectively measured by ward nurses blinded to the NSCAN variables. A database was created to pair the RASS scores and activity scores at night on the same day. ResultsDuring the observation period, 1209 patients were hospitalized, and 3028 pairs of assessments of 971 patients were analyzed. The NSCAN activity scores significantly increased with increasing RASS scores (Jonckheere–Terpstra test, p < 0.001). The mean values of the activity scores for each RASS score were RASS −5, 28.9; RASS −4, 36.4; RASS −3, 41.7; RASS −2, 57.4; RASS −1, 58.8; RASS 0, 62.6; RASS 1, 79.6; RASS 2, 106.5; and RASS 3, 118.7. ConclusionThe NSCAN activity significantly correlated with modified RASS agitation scores. Real-time NSCAN data on agitation may aid timely interventions for optimal symptom control. To improve outcomes for patients suffering from terminal delirium, more research on monitoring tools is warranted.