analiza aiqbench
SMARtCARE introduces a privacy-preserving architecture for clinical AI that mitigates long-context limitations by using trajectory fingerprints to flag potential patient deterioration without automatically exposing full medical records. The system enforces a four-state escalation workflow where full data retrieval requires explicit clinician action, ensuring auditability and preventing unauthorized access to sensitive history. While the study validates logic stability on MIMIC datasets rather than clinical outcomes, it offers a significant methodological advance for deploying agentic AI in high-stakes environments like ICUs.
SMARtCARE: Privacy-Preserving Agentic AI Systems for Bounded-Autonomy Clinical Decision Support
arXiv:2609.31763v1 Announce Type: new Abstract: Long-context clinical AI systems can miss relevant patient history when prior admissions fall outside the active reasoning context. In ICU monitoring, this can cause early vital-sign drift to appear nonspecific even when it resembles a prior deterioration pattern. SMARtCARE addresses this gap through a four-state clinical decision-support architecture: Stable, Meta-cognitive, Assisted, and Regulated (Revoked). Rather than automatically retrieving prior records, SMARtCARE uses a lossy six-channel fingerprint of the patient's prior trajectory. When current drift matches that fingerprint and the prior record is absent from context, the system raises a Meta-cognitive escalation for clinician review; full retrieval occurs only through clinician action in the Assisted state. A patient-identity guard is designed to enforce correct attribution across data loading, logging, and audit layers. Evaluation combines a synthetic Monte Carlo study that validates the state-transition logic and estimator stability, not clinical performance, with real-data runs on both the MIMIC-III and MIMIC-IV Clinical Database Demos. On MIMIC-III, one prior-pattern recurrence was identified among 14 two-admission patients; on MIMIC-IV, the same pipeline produced no fingerprint matches among 9 two-admission patients, which illustrates a key limitation of a fixed canonical pattern library. Across both runs all logged decisions were fully traceable and correctly attributed. The results support SMARtCARE as a traceable, privacy-aware mechanism for surfacing middle-context risk; they are not a clinical efficacy claim.
Abstract:Long-context clinical AI systems can miss relevant patient history when prior admissions fall outside the active reasoning context. In ICU monitoring, this can cause early vital-sign drift to appear nonspecific even when it resembles a prior deterioration pattern. SMARtCARE addresses this gap through a four-state clinical decision-support architecture: Stable, Meta-cognitive, Assisted, and Regulated (Revoked). Rather than automatically retrieving prior records, SMARtCARE uses a lossy six-channel fingerprint of the patient's prior trajectory. When current drift matches that fingerprint and the prior record is absent from context, the system raises a Meta-cognitive escalation for clinician review; full retrieval occurs only through clinician action in the Assisted state. A patient-identity guard is designed to enforce correct attribution across data loading, logging, and audit layers. Evaluation combines a synthetic Monte Carlo study that validates the state-transition logic and estimator stability, not clinical performance, with real-data runs on both the MIMIC-III and MIMIC-IV Clinical Database Demos. On MIMIC-III, one prior-pattern recurrence was identified among 14 two-admission patients; on MIMIC-IV, the same pipeline produced no fingerprint matches among 9 two-admission patients, which illustrates a key limitation of a fixed canonical pattern library. Across both runs all logged decisions were fully traceable and correctly attributed. The results support SMARtCARE as a traceable, privacy-aware mechanism for surfacing middle-context risk; they are not a clinical efficacy claim.
Artificial Intelligence (cs.AI); Emerging Technologies (cs.ET); Software Engineering (cs.SE)
arXiv:2609.31763 [cs.AI]
(or arXiv:2609.31763v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.31763 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 318 cuvinte
Subiecte conexe
Intelligence as Managed Autonomy: Failure, Escalation, and Governance for Agentic AI Systems
arXiv · 12 iun. 2026
Biomedical Imaging's Real Bottleneck Is the Data, Not the Model
Databricks · 8 oct. 2026
HiMA-MDD: A Hierarchical Multi-Agent Harness for Interpretable Multimodal Depression Detection in Clinical Interviews
arXiv · 25 aug. 2026
How Telcos Build Autonomous Networks with Agentic AI
NVIDIA · 23 iun. 2026