Deployer-side XAI instrumentation for regulated AI: a clinical case study in ICL sizing
Cargando...
Fecha
2026
Título de la revista
ISSN de la revista
Título del volumen
Editor
Science and Information Organization
Resumen
Regulated AI creates a monitoring problem for deployers who must organise human oversight, log-retention and post-market surveillance while often having access only to the prediction interface. This study specifies a deployer-side XAI instrumentation protocol for the output→action boundary, where a model output becomes a reason for action. The protocol reorganises KernelSHAP, nearest-neighbour envelope checks, bounded perturbation, and rank-order stability into a per-decision evidence record computed from predict() calls. We instantiate the protocol in a clinical case of phakic Implantable Collamer Lens sizing, using a 55-eye held-out cohort and an Extra Trees regressor for post-operative vault prediction. The record contains five signals: score_margin, constraint_enforcement, envelope_validity, decision_robustness, and record_integrity, plus two cohort-level oversight aggregates. The case study shows how the same record can support decision-time human oversight, later audit and post-market surveillance under the EU AI Act and the Medical Devices Regulation.
Palabras clave
EU AI Act
Explainable AI
Human oversight
ICL sizing
Medical Device Software
XAI instrumentation
Explainable AI
Human oversight
ICL sizing
Medical Device Software
XAI instrumentation
Descripción
Materias
Cita
Urrutia-Onate, D., Onieva, E., & Perallos, A. (2026). Deployer-side XAI instrumentation for regulated AI: a clinical case study in ICL sizing. International Journal of Advanced Computer Science and Applications, 17(6), 45-61. https://doi.org/10.14569/IJACSA.2026.0170605