A Principled Framework for Safe Algorithm Updates in Automated Insulin Delivery Systems

• URL: http://arxiv.org/abs/2606.13882v1
• Date: Thu, 11 Jun 2026 20:18:13 GMT
• Title: A Principled Framework for Safe Algorithm Updates in Automated Insulin Delivery Systems
• Abstract summary: Our framework classifies bugs and evaluates the clinical equivalence of AID system software updates.<n>It is system-agnostic and applicable to all widely used OS-AID systems.
• Score: 0.1915630210833957
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Background: AID algorithms require ongoing software updates and bug fixes. In co-adapted systems, where users tune settings around existing algorithmic behavior, bug fixes can paradoxically disrupt glycemic control. No principled framework evaluates the safety of AID algorithm updates. Methods: Our two-part framework classifies bugs and evaluates the clinical equivalence of AID system software updates. Bugs are classified as factual, heuristic, or computational, each with distinct management strategies. Classifications were validated from porting Trio's oref algorithm from Javascript to a bug-fixed Swift implementation. We compared implementations using shadow execution on 736,480 invocations from eight Trio users. The second component assesses clinical equivalence with error analysis on paired glucose values, applied to both Trio implementations using mechanistic in silico and data-driven replay simulation. Results: In mechanistic in silico simulation, the Swift and Javascript implementations produced nearly identical Time in Range (84.9% vs. 84.9%) and Glycemia Risk Index (23.5% vs. 23.9%), with more than 99% of paired glucose in Parkes Error Grid Zones A and B, meeting our clinical equivalence threshold. Shadow execution showed low mismatch rates in oref components (iob 0.43%, autosens 1.22%, determineBasal 0.07%, meal 0.01%), with clinically meaningful differences in 0.03% of iob invocations. Data-driven replay simulations of bugs revealed more than 99% of downstream paired glucose in Parkes Error Grid Zones A and B, also meeting our clinical equivalence threshold. Conclusions: Our framework integrates bug-fixing principles with multi-method clinical evaluation to assess AID algorithm update safety. It is system-agnostic and applicable to all widely used OS-AID systems, with case studies highlighting the need for systematic remediation of factual and computational bugs.

Related papers

• Not All Relations Are Equal: Relation-Balanced and Calibrated Graph Learning for Provenance-Based Intrusion Detection [66.44527094471619]
  We present RECAL, an unsupervised framework using relation-balanced masked graph learning to better capture rare interaction patterns.<n>On three DARPA E3 datasets, RECAL achieves F1 scores of 99.99%, 99.93%, and 99.99%, outperforming the best baseline on each dataset by 0.88, 0.82, and 0.42 percentage points, respectively.
  arXiv  Detail & Related papers  (2026-09-15T00:31:01Z)
• AgentEval: DAG-Structured Step-Level Evaluation for Agentic Workflows with Error Propagation Tracking [13.891522069967507]
  We present AgentEval, a framework that formalizes agent executions as evaluation directed acyclic graphs (DAGs)<n>AgentEval achieves 2.17x higher failure detection recall than end-to-end evaluation and 72% root cause accuracy against an 81% human ceiling.
  arXiv  Detail & Related papers  (2026-04-26T07:38:47Z)
• ImplicitMemBench: Measuring Unconscious Behavioral Adaptation in Large Language Models [60.14219417402433]
  Existing memory benchmarks for LLM agents evaluate explicit recall of facts, yet overlook implicit memory where experience becomes automated behavior without conscious retrieval.<n>We introduce ImplicitMemBench, the first systematic benchmark evaluating implicit memory through three constructs drawn from standard cognitive-science accounts of non-declarative memory.<n>Our 300-item suite employs a unified Learning/Priming-Interfere-Test protocol with first-attempt scoring.
  arXiv  Detail & Related papers  (2026-04-09T10:26:32Z)
• TRAJEVAL: Decomposing Code Agent Trajectories for Fine-Grained Diagnosis [23.834704102474927]
  Code agents can resolve GitHub issues, yet when they fail, current evaluation provides no visibility into where or why.<n>We introduce TRAJEVAL, a diagnostic framework that decomposes agent trajectories into three interpretable stages.<n>We validate that these diagnostics are predictive, achieving model-level Pass@1 prediction within 0.87-2.1% MAE.
  arXiv  Detail & Related papers  (2026-03-25T05:27:03Z)
• Optimization Instability in Autonomous Agentic Workflows for Clinical Symptom Detection [3.0950658457067433]
  We investigate a phenomenon in which continued autonomous improvement paradoxically degrades classifier performance.<n>We observed that validation sensitivity oscillated between 1.0 and 0.0 across iterations, with severity inversely proportional to class prevalence.<n>With selector agent oversight, the system outperformed expert-curated lexicons on brain fog detection by 331% (F1) and chest pain by 7%, despite requiring only a single natural language term as input.
  arXiv  Detail & Related papers  (2026-02-17T21:45:20Z)
• On Randomness in Agentic Evals [6.177270420667714]
  Agentic systems are evaluated on benchmarks where agents interact with environments to solve tasks.<n>Most papers report a pass@1 score computed from a single run per task, assuming this gives a reliable performance estimate.<n>We find substantial variance: single-run pass@1 estimates vary by 2.2 to 6.0 percentage points depending on which run is selected.
  arXiv  Detail & Related papers  (2026-02-06T19:49:13Z)
• TrustJudge: Inconsistencies of LLM-as-a-Judge and How to Alleviate Them [58.04324690859212]
  Large Language Models (LLMs) as automated evaluators (LLM-as-a-judge) has revealed critical inconsistencies in current evaluation frameworks.<n>We identify two fundamental types of inconsistencies: Score-Comparison Inconsistency and Pairwise Transitivity Inconsistency.<n>We propose TrustJudge, a probabilistic framework that addresses these limitations through two key innovations.
  arXiv  Detail & Related papers  (2025-09-25T13:04:29Z)
• Adaptive Monitoring and Real-World Evaluation of Agentic AI Systems [3.215065407261898]
  Multi-agent systems that combine large language models with external tools are rapidly transitioning from research laboratories into high-stakes domains.<n>This "Advanced" sequel fills that gap by providing an algorithmic instantiation or empirical evidence.<n>AMDM cuts anomaly-detection latency from 12.3 s to 5.6 s on simulated goal drift and reduces false-positive rates from 4.5% to 0.9%.
  arXiv  Detail & Related papers  (2025-08-28T15:52:49Z)
• Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solving [62.71545696485824]
  We introduce AGENT KB, a universal memory infrastructure enabling seamless experience sharing across heterogeneous agent frameworks without retraining.<n>AGENT KB aggregates trajectories into a structured knowledge base and serves lightweight APIs.<n>We validate AGENT across major frameworks on GAIA, Humanity's Last Exam, GPQA, and SWE-bench.
  arXiv  Detail & Related papers  (2025-07-08T17:59:22Z)
• Improving Bias Correction Standards by Quantifying its Effects on Treatment Outcomes [54.18828236350544]
  Propensity score matching (PSM) addresses selection biases by selecting comparable populations for analysis. Different matching methods can produce significantly different Average Treatment Effects (ATE) for the same task, even when meeting all validation criteria. To address this issue, we introduce a novel metric, A2A, to reduce the number of valid matches.
  arXiv  Detail & Related papers  (2024-07-20T12:42:24Z)
• Quality-Based Conditional Processing in Multi-Biometrics: Application to Sensor Interoperability [63.05238390013457]
  We describe and evaluate the ATVS-UAM fusion approach submitted to the quality-based evaluation of the 2007 BioSecure Multimodal Evaluation Campaign. Our approach is based on linear logistic regression, in which fused scores tend to be log-likelihood-ratios. Results show that the proposed approach outperforms all the rule-based fusion schemes.
  arXiv  Detail & Related papers  (2022-11-24T12:11:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.