---
title: "Model-directed insulin pump dosing for type 2 diabetes using LSTM: development and prospective eva"
id: "medrxiv-22-a-data-driven-system-for-insulin-pump-dosing-in-type-2-diabetes-development"
canonical_url: "https://medichelpline.com/clinical-feed/medrxiv-22-a-data-driven-system-for-insulin-pump-dosing-in-type-2-diabetes-development"
content_type: "clinical_feed_article"
specialty: "Endocrinology"
source_name: "medRxiv (Clinical Preprints)"
source_url: "https://www.medrxiv.org/content/10.64898/2026.10.13.26365400v1?rss=1"
published_at: "2026-10-15T12:00:00.000Z"
evidence_level: "Verified Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Model-directed insulin pump dosing for type 2 diabetes using LSTM: development and prospective eva
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/medrxiv-22-a-data-driven-system-for-insulin-pump-dosing-in-type-2-diabetes-development
- **Specialty:** [Endocrinology](https://medichelpline.com/clinical-feed/endocrinology.md)
- **Primary Source:** medRxiv (Clinical Preprints)
- **Source URL:** [Original Journal Publication](https://www.medrxiv.org/content/10.64898/2026.10.13.26365400v1?rss=1)
- **Published At:** 2026-10-15T12:00:00.000Z
- **Evidence Rating:** Verified Feed
## Executive GIST (TL;DR)
- The authors developed a dosing system for open-loop **insulin pump** therapy using a long short-term memory (LSTM) model trained on 1,738 treatment courses from 1,641 patients with **type 2 diabetes**. - The system provides recommendations for initial prescribing and daily basal and bolus titration; development used patient-grouped five-fold validation to assess agreement with recorded prescriptions. - Validation showed closer concordance with clinician prescriptions for subsequent titration decisions than for initial dose initiation, according to the development cohort performance reported. - The team performed a prospective evaluation within the model-directed arm of a randomized trial that included 97 patients; median derived **time in range** (3.9–10.0 mmol/L) was 84.5% in that cohort, with low hypoglycemic exposure reported. - Favorable glycemic control was observed even among patients who had prior insulin treatment before study entry. - Higher mean glucose during the first 24 hours correlated with worse glycemic control later in treatment, indicating early response is a relevant signal. - The authors examined the relationship between a patient’s distance from the development dataset and glycemic outcomes; that association strengthened when early glucose responses were included alongside baseline characteristics. - The study team noted a pending Chinese patent application related to the AI-based dosing method and declared no other competing interests. - Trial registration is provided (ClinicalTrials.gov NCT07066891). The manuscript is a preprint and has not undergone peer review; datasets are not publicly available but de-identified data may be requested from the corresponding author after peer-reviewed publication.
## Clinical Analysis & Structured Key Points
A data-driven system for insulin pump dosing in type 2 diabetes: development, prospective evaluation and clinical applicability | medRxiv Skip to main content A data-driven system for insulin pump dosing in type 2 diabetes: development, prospective evaluation and clinical applicability View ORCID Profile Lewen Cai , Shuhong Huang , Yuping Cao , View ORCID Profile Zhimin Huang doi: https://doi.org/10.64898/2026.10.13.26365400 Lewen Cai 1 University of Technology Sydney; Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Lewen Cai Shuhong Huang 2 Guangzhou Hong-tech Internet Service Co., Ltd.; Find this author on Google Scholar Find this author on PubMed Search for this author on this site Yuping Cao 3 Chaozhou Central Hospital; Find this author on Google Scholar Find this author on PubMed Search for this author on this site Zhimin Huang 4 The First Affiliated Hospital of Sun Yat-sen University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Zhimin Huang For correspondence: hzhim{at}mail.sysu.edu.cn Abstract Info/History Metrics Supplementary material Data/Code Preview PDF Abstract Open-loop insulin pump therapy depends on clinical expertise for dose initiation and titration. We developed a long short-term memory-based system, using 1738 treatment courses from 1641 patients with type 2 diabetes for initial prescribing and daily basal and bolus titration. Patient-grouped five-fold validation showed closer agreement with recorded prescriptions for subsequent titration than for initiation. Prospective evaluation in 97 patients from the model-directed arm of a randomized trial yielded a median derived time in range (3.9-10.0mmol/L) of 84.5%, with low hypoglycemic exposure. Favorable control extended to previously treated patients. Higher mean glucose during the first 24 hours was associated with poorer glycemic control thereafter. The association between distance from development data and glycemic outcomes was stronger when early glucose responses were included alongside baseline characteristics. These findings support model-directed pump dosing and highlight early treatment response as a clinically relevant signal for assessing applicability and identifying patients who may require additional oversight. Trial registration: ClinicalTrials.gov, NCT07066891 . Competing Interest Statement Z.H., S.H. and Y.C. are named inventors on a pending Chinese patent application (no. 202510907499.2), filed by the First Affiliated Hospital of Sun Yat-sen University, concerning the AI-based insulin pump dosing method described in this study. The authors declare no other competing interests. Clinical Trial NCT07066891 Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: The study was approved by the ethics board of the First Affiliated Hospital of Sun Yat-sen University ([2024]-836), and all participants in RCT provided written informed consent. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Data Availability The datasets generated and analysed in this study are not publicly available to protect patient privacy. De-identified data supporting the findings will be available from the corresponding author upon reasonable request following publication in a peer-reviewed journal, subject to applicable ethical and institutional requirements. Funder Information Declared National Natural Science Foundation of China , 82070918 Copyright The copyright holder for this preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license . Back to top Previous Next Posted October 15, 2026. Download PDF Supplementary Material Data/Code Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following A data-driven system for insulin pump dosing in type 2 diabetes: development, prospective evaluation and clinical applicability Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share A data-driven system for insulin pump dosing in type 2 diabetes: development, prospective evaluation and clinical applicability Lewen Cai , Shuhong Huang , Yuping Cao , Zhimin Huang medRxiv 2026.10.13.26365400; doi: https://doi.org/10.64898/2026.10.13.26365400 Share This Article: Copy Citation Tools A data-driven system for insulin pump dosing in type 2 diabetes: development, prospective evaluation and clinical applicability Lewen Cai , Shuhong Huang , Yuping Cao , Zhimin Huang medRxiv 2026.10.13.26365400; doi: https://doi.org/10.64898/2026.10.13.26365400 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Areas All Articles Addiction Medicine (629) Allergy and Immunology (914) Anesthesia (343) Cardiovascular Medicine (4924) Dentistry and Oral Medicine (488) Dermatology (421) Emergency Medicine (667) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1655) Epidemiology (16117) Forensic Medicine (35) Gastroenterology (1228) Genetic and Genomic Medicine (7169) Geriatric Medicine (740) Health Economics (1080) Health Informatics (5181) Health Policy (1452) Health Systems and Quality Improvement (1808) Hematology (596) HIV/AIDS (1363) Infectious Diseases (except HIV/AIDS) (16398) Intensive Care and Critical Care Medicine (1187) Medical Education (685) Medical Ethics (156) Nephrology (733) Neurology (7436) Nursing (372) Nutrition (1111) Obstetrics and Gynecology (1274) Occupational and Environmental Health (1020) Oncology (3690) Ophthalmology (1077) Orthopedics (400) Otolaryngology (461) Pain Medicine (488) Palliative Medicine (142) Pathology (714) Pediatrics (1837) Pharmacology and Therapeutics (744) Primary Care Research (780) Psychiatry and Clinical Psychology (6035) Public and Global Health (9885) Radiology and Imaging (2479) Rehabilitation Medicine and Physical Therapy (1495) Respiratory Medicine (1258) Rheumatology (654) Sexual and Reproductive Health (791) Sports Medicine (591) Surgery (794) Toxicology (109) Transplantation (313) Urology (299)
## Related Clinical Research

- [Olive oil before carbs and blood sugar: what the mouse study found](https://medichelpline.com/clinical-feed/medical-news-today-0-olive-oil-shot-trend-what-science-really-says-about-fat-and-glucose-response.md)
- [Diabetic retinopathy regression rates in very mild and mild NPDR in an IHS teleophthalmology cohort](https://medichelpline.com/clinical-feed/plos-one-2-rates-of-nonproliferative-diabetic-retinopathy-npdr-regression-among-patients.md)
- [Cardiovascular safety of oral orforglipron versus insulin glargine in T2D at increased CV risk (AC](https://medichelpline.com/clinical-feed/pubmed-42815506.md) (DOI: 10.1016/S0140-6736(26)01865-9)
- [Self-Management Program Lowers HbA1c, eGFR, SBP, LDL-C in High-Risk DKD Patients with Type 2 Diabe](https://medichelpline.com/clinical-feed/plos-one-11-effectiveness-of-a-self-management-intervention-on-clinical-outcomes-among.md)
- [Endocrine Society Comments on the 2027 Medicare Physician Fee Schedule Proposed Rule](https://medichelpline.com/clinical-feed/endocrine-news-0-society-submits-comments-on-the-medicare-physician-fee-schedule-proposed-rule.md)

## Navigation
- [← Back to Endocrinology Feed](https://medichelpline.com/clinical-feed/endocrinology.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.