Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

This study introduces Glucose Level Understanding and Control Optimized for Safety and Efficacy (GLUCOSE), a distributional offline reinforcement learning algorithm for optimizing insulin dosing after cardiac surgery. Trained on 5228 patients, tested on 920, and externally validated on 649, GLUCOSE achieved a mean estimated reward of 0.0 [-0.07, 0.06] in internal testing and -0.63 [-0.74, -0.52] in external validation, outperforming clinician returns of -1.29 [-1.37, -1.20] and -1.02 [-1.16, -0.89]. In multi-phase human validation, GLUCOSE first showed a significantly lower mean absolute error (MAE) in insulin dosing, with 0.9 units MAE versus clinicians' 1.97 units (p < 0.001) in internal testing and 1.90 versus 2.24 units (p = 0.003) in external validation. The second and third phases found GLUCOSE's performance as comparable to or exceeding that of senior clinicians in MAE, safety, effectiveness, and acceptability. These findings suggest GLUCOSE as a robust tool for improving postoperative glucose management.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12116759PMC
http://dx.doi.org/10.1038/s41746-025-01709-9DOI Listing

Publication Analysis

Top Keywords

reinforcement learning
8
cardiac surgery
8
insulin dosing
8
glucose
5
distributional reinforcement
4
learning model
4
model optimal
4
optimal glucose
4
glucose control
4
control cardiac
4

Similar Publications

Background: Primary liver cancer, particularly hepatocellular carcinoma (HCC), poses significant clinical challenges due to late-stage diagnosis, tumor heterogeneity, and rapidly evolving therapeutic strategies. While systematic reviews and meta-analyses are essential for updating clinical guidelines, their labor-intensive nature limits timely evidence synthesis.

Objective: This study proposes an automated literature screening workflow powered by large language models (LLMs) to accelerate evidence synthesis for HCC treatment guidelines.

View Article and Find Full Text PDF

Distinct prelimbic cortex ensembles encode response execution and inhibition.

Proc Natl Acad Sci U S A

September 2025

Behavioral Neuroscience Research Branch, Intramural Research Program, National Institute on Drug Abuse, Baltimore, MD 21224.

Learning when to initiate or withhold actions is essential for survival, requiring the integration of past experiences with new information to adapt to changing environments. The prelimbic cortex (PL) plays a central role in this process, with a stable PL neuronal population (ensemble) recruited during operant reward learning to encode response execution. However, it is unknown how this established reward-learning ensemble adapts to changing reward contingencies, such as reward omission during extinction.

View Article and Find Full Text PDF

Cervical cancer remains a significant cause of female mortality worldwide, primarily due to abnormal cell growth in the cervix. This study proposes an automated classification method to enhance detection accuracy and efficiency, addressing contrast and noise issues in traditional diagnostic approaches. The impact of image enhancement on classification performance is evaluated by comparing transfer learning-based Convolutional Neural Network (CNN) models trained on both original and enhanced images.

View Article and Find Full Text PDF

Previous studies suggested that acute stress can impair flexible goal-directed action control in favor of habitual action control. In addition, there is evidence that acute stress differentially affects the processing of rewards and punishments. Therefore, we aimed at investigating whether acute stress affects the balance between goal-directed and habitual behavior not only for behavior aiming at reward but also for behavior motivated by avoiding punishments.

View Article and Find Full Text PDF

Artificial intelligence (AI) based anticancer drug recommendation systems have emerged as powerful tools for precision dosing. Although existing methods have advanced in terms of predictive accuracy, they encounter three significant obstacles, including the "black-box" problem resulting in unexplainable reasoning, the computational difficulty for graphbased structures, and the combinatorial explosion during multistep reasoning. To tackle these issues, we introduce a novel Macro-Micro agent Drug sensitivity inference (MarMirDrug).

View Article and Find Full Text PDF