Artificial Intelligence replication of District Hospital Sickle Cell Crisis Triage Under Unmeasured Capillary Refill Videography Motion Artefact: A Healthcare Sensitivity-Stratified Safety Evaluation

Capillary refill time (CRT) videography is increasingly proposed as a triage adjunct for sickle cell crisis in district hospitals, yet the motion artefact that contaminates handheld recordings is rarely measured and never reported alongside the triage decision it informs. the present analysis treats that unmeasured artefact as a sensitivity parameter and asks how the safety of an artificial-intelligence triage rule degrades as the artefact distribution widens. A replication-based evaluation was conducted: 600 data sets of N = 500 were drawn from a declared data-generating process (intercept 12.0, exposure 0.42, group -3.1, severity 1.35, error SD 4.2), and the behaviour of the ordinary least squares estimator was characterised under the stated process. Across the 600 replications the intercept bias was 0.01411 and the exposure bias was -0.00034; the root mean squared errors were 0.8687 and 0.01552 respectively; empirical coverage of the nominal 95% confidence intervals was 1.0 for both parameters. The sampling distribution of the exposure coefficient was approximately normal and the Q-Q plot was close to the identity line. A receiver operating characteristic constructed from the same replications located the triage threshold where sensitivity and false-positive rate trade off. The contribution is a sensitivity-stratified safety argument: because the artefact is unmeasured, the only defensible triage policy is one whose threshold is chosen against an explicit artefact-widening scenario rather than against the nominal calibration. The paper specifies that scenario and states the conditions under which the rule should be suspended.