The paper presents a reliability‑aware benchmarking framework for short‑term glycemic risk prediction, using 5,503,441 prediction anchors from 736 type‑1 diabetes patients. Three neural models—a Hybrid LSTM‑Transformer, a Multimodal Digital Twin, and a State‑Updating Digital Twin—were evaluated on 30‑, 60‑, and 120‑minute horizons for hypoglycemia and hyperglycemia, with identical patient‑level validation. While no significant differences in predictive accuracy were found, the models displayed distinct reliability traits: the Multimodal Digital Twin had the lowest Brier score, the State‑Updating Digital Twin the lowest expected calibration error, and the Hybrid LSTM‑Transformer the strongest explanation faithfulness.
By Pir Bakhsh Khokhar, Carmine Gravino, Fabio Palomba
FairGlucose is a 300‑patient CGM cohort balanced across 12 demographic strata, providing 132,480 forecasting samples and 3,945 behavioral events. Benchmarking 33 models on 2‑hour glucose forecasting revealed that population‑level validation masks significant subgroup disparities, with error ratios ranging from 0.8 to 1.4 and T1D patients experiencing 6 mg/dL higher error than T2D. The study shows that these gaps persist across all models, align with clinically hard cases, and vary with input‑length sensitivity, underscoring the need for subgroup‑disaggregated reporting in digital health AI.
By Junjie Luo, Xuzhe Zhi, Rui Han, Abhimanyu Kumbara, Anand K. Iyer, Mansur E. Shomali, Ritu Agarwal, Guodong Gordon Gao
The study examines how temporal resampling (binning) of irregular clinical data affects the safety and performance of offline reinforcement learning (ORL) models in healthcare. By analyzing MIMIC‑IV insulin infusion records, the authors show that binning can reverse causal order and obscure critical events. In simulated trials with virtual type‑1 diabetes patients, ORL models trained on binned data performed up to 60% worse than those trained on unprocessed data, yet retrospective evaluation on resampled data overestimated their returns.
By Thomas Frost, Hrisheekesh Vaidya, Steve Harris
The study introduces PMB‑NN, a physiological‑model‑based neural network that forecasts glucose levels in type 1 diabetes using a temporal convolutional network trained on 180 minutes of historical data and constrained by a personalized four‑parameter physiological model. Evaluated on 6–8 weeks of continuous glucose monitoring, pump, and meal data from 25 participants, PMB‑NN achieved median mean absolute errors of 16.24, 25.00, and 30.20 mg/dL for 30‑, 60‑, and 120‑minute horizons when basal insulin records were available, outperforming the physiological model alone across all settings. The model also demonstrated expected directional changes for most correction boluses and carbohydrate rescues, and its three‑term variant improved hyperglycaemia accuracy with basal records without significantly altering overall error.
By Yaowen Zhang, Shuhao Que, Ying Wang
The paper introduces a framework that distinguishes two causes of saturation in clinical prediction: a learner gap, where the model fails to use available information, and a measurement‑channel ceiling, where the recorded variables limit performance. It provides theoretical characterizations, finite‑sample diagnostics, and empirical audits across three large cohorts, showing that well‑tuned models approach the frontier while deficient learners leave large gaps. A PRISMA‑guided synthesis across 104 tasks reveals consistent channel‑level patterns, suggesting that improving the learner or the measurement channel can audit and potentially lift performance.
By Sayeed Shafayet Chowdhury, Nusrat Jahan, Snehasis Mukhopadhyay, Shiaofen Fang, Vijay R. Ramakrishnan
arXiv:2606. 18640v1 Announce Type: new Abstract: Glucose forecasting algorithms are an important aspect of glycemic control management in type 1 diabetes.
By Nathaniel Jeffries, Miriam Wolff, Sam Royston, Elizabeth Healey, Caleb Mayer, David Klonoff, Michael Snyder, Tao Wang
arXiv:2606. 03198v1 Announce Type: cross Abstract: Clinical AI evaluation increasingly delegates scoring to large language models (LLMs) acting as AI raters, yet their scoring behavior across evaluation conditions has not been quantitatively characterized.
By Sangwon Baek, Kyu Yeon Hur, Kyunga Kim
arXiv:2606. 29386v1 Announce Type: new Abstract: Predicting a patient's physiological trajectory under a planned treatment sequence is a prospective interventional problem, not standard time-series extrapolation.
By Amirreza Dolatpour Fathkouhi, Justin Lee, Heman Shakeri
arXiv:2606. 24145v1 Announce Type: new Abstract: Large language models (LLMs) can produce clinically fluent recommendations for type 2 diabetes while failing to satisfy guideline constraints or explicitly justify lifestyle-related glycemic claims.
By Saba A. Farahani, Hung Cao, Ramesh Jain, Amir M. Rahmani
arXiv:2606. 06881v1 Announce Type: new Abstract: Blood glucose forecasting models are foundational for modern diabetes management systems, as reliable short-term predictions can enable proactive interventions, support automated insulin delivery, and reduce the risk of hypo- and hyperglycemic events.
By Baiying Lu, Zhaohui Liang, Ryan Pontius, Shengpu Tang, Temiloluwa Prioleau
arXiv:2604. 23954v2 Announce Type: replace Abstract: Artificial Intelligence (AI) and Machine Learning (ML) models used in clinical settings are increasingly deployed to support clinical decision-making.
By Ioannis Bilionis, Ricardo C. Berrios, Luis Fernandez-Luque, Carlos Castillo
arXiv:2607. 16253v1 Announce Type: cross Abstract: Machine learning-based Type 2 diabetes risk prediction models obtain good internal validation results but lose effectiveness in real-world applications due to deficient external testing and fairness assessment.
By Rajveer Singh Pall, Sameer Yadav, Siddharth Bhalerao, Sourabh Sahu, Ritu Ahluwalia, Bhaskar Awadhiya