{"id":48513058,"url":"https://github.com/samankhamesian/guide","last_synced_at":"2026-04-07T18:01:33.463Z","repository":{"id":348776936,"uuid":"1005887007","full_name":"SamanKhamesian/GUIDE","owner":"SamanKhamesian","description":"We present GUIDE (Glucose Intelligence Decision Engine), an RL-based framework that generates patient-specific recommendations for insulin and carbohydrate intake, adapting to individual glucose dynamics and daily routines ","archived":false,"fork":false,"pushed_at":"2026-04-02T19:24:28.000Z","size":139,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-04-03T06:38:51.897Z","etag":null,"topics":["automated-insulin-delivery","continuous-glucose-monitoring","cql-algorithm","ppo","reinforcement-learning","sac","simulation","td3-bc","type-1-diabetes"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/SamanKhamesian.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-06-21T02:49:50.000Z","updated_at":"2026-04-02T19:24:31.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/SamanKhamesian/GUIDE","commit_stats":null,"previous_names":["samankhamesian/guide"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/SamanKhamesian/GUIDE","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SamanKhamesian%2FGUIDE","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SamanKhamesian%2FGUIDE/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SamanKhamesian%2FGUIDE/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SamanKhamesian%2FGUIDE/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SamanKhamesian","download_url":"https://codeload.github.com/SamanKhamesian/GUIDE/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SamanKhamesian%2FGUIDE/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31522574,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-07T16:28:08.000Z","status":"ssl_error","status_checked_at":"2026-04-07T16:28:06.951Z","response_time":105,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["automated-insulin-delivery","continuous-glucose-monitoring","cql-algorithm","ppo","reinforcement-learning","sac","simulation","td3-bc","type-1-diabetes"],"created_at":"2026-04-07T18:01:31.236Z","updated_at":"2026-04-07T18:01:33.368Z","avatar_url":"https://github.com/SamanKhamesian.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# GUIDE\n\n**GUIDE: Reinforcement Learning for Behavioral Action Support in Type 1 Diabetes**\n\nType 1 Diabetes (T1D) management requires continuous adjustment of insulin and lifestyle behaviors to maintain blood glucose within a safe target range. \nAlthough automated insulin delivery (AID) systems have improved glycemic outcomes, many patients still fail to achieve recommended clinical targets, warranting new approaches to improve glucose control in patients with T1D. \nWhile reinforcement learning (RL) has been utilized as a promising approach, current RL-based methods focus primarily on insulin-only treatment and do not provide behavioral recommendations for glucose control. \nTo address this gap, we propose GUIDE, an RL-based decision-support framework designed to complement AID technologies by providing behavioral recommendations to prevent abnormal glucose events. \nGUIDE generates structured actions defined by intervention type, magnitude, and timing, including bolus insulin administration and carbohydrate intake events. \nGUIDE integrates a patient-specific glucose level predictor trained on real-world continuous glucose monitoring data and supports both offline and online RL algorithms within a unified environment. \nWe evaluate both off-policy and on-policy methods across 25 individuals with T1D using standardized glycemic metrics. Among the evaluated approaches, the CQL-BC algorithm demonstrates the highest average time-in-range, reaching 85.49% while maintaining low hypoglycemia exposures. Behavioral similarity analysis further indicates that the learned CQL-BC policy preserves key structural characteristics of patient action patterns, achieving a mean cosine similarity of 0.87 ± 0.09 across subjects. \nThese findings suggest that conservative offline RL with a structured behavioral action space can provide clinically meaningful and behaviorally plausible decision support for personalized diabetes management.\n\n---\n\n## 📁 Dataset Setup\n\n### AZT1D Dataset\n\n- **Download from Mendeley:**  \n  https://data.mendeley.com/datasets/gk9m674wcx/1\n\n- **Directory structure:**  \n  Place ```AZT1D``` folder in the ```./GUIDE/dataset/``` directory:\n\n---\n\n## ⚙️ Environment Setup\n\n- **Python version:** `3.12`\n\n- **Install dependencies:**\n\n  Create a virtual environment (optional but recommended):\n\n  ```bash\n  python -m venv venv\n  source venv/bin/activate  # On Windows: venv\\Scripts\\activate\n  ```\n\n  Then install required packages:\n\n  ```bash\n  pip install -r requirements.txt\n  ```\n\n  `requirements.txt` includes:\n\n  ```\n  matplotlib==3.10.8\n  numpy==2.4.2\n  pandas==3.0.1\n  scikit_learn==1.8.0\n  scipy==1.17.1\n  statsmodels==0.14.6\n  tensorflow==2.20.0\n  torch==2.7.0\n  ```\n---\n\n## 📖 Citation\n\nIf you use GLIMMER in your work, please cite:\n\n```bibtex\n@misc{khamesian2026guidereinforcementlearningbehavioral,\n      title={GUIDE: Reinforcement Learning for Behavioral Action Support in Type 1 Diabetes}, \n      author={Saman Khamesian and Sri Harini Balaji and Di Yang Shi and Stephanie M. Carpenter and Daniel E. Rivera and W. Bradley Knox and Peter Stone and Hassan Ghasemzadeh},\n      year={2026},\n      eprint={2604.00385},\n      archivePrefix={arXiv},\n      primaryClass={cs.LG},\n      url={https://arxiv.org/abs/2604.00385}, \n}\n```\n\n---\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsamankhamesian%2Fguide","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsamankhamesian%2Fguide","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsamankhamesian%2Fguide/lists"}