Insulin Titration System Based on Deep Learning

NCT ID: NCT05409391

Last Updated: 2023-06-07

Basic Information

• Recruitment Status: COMPLETED
• Clinical Phase: NA
• Total Enrollment: 16 participants
• Study Classification: INTERVENTIONAL
• Study Start Date: 2022-06-15
• Study Completion Date: 2022-10-06

Brief Summary

This is an open-labeled, one-arm intervention trial to access the effect and safety of the Insulin Titration System Based on Deep Learning in patients with Type 2 Diabetes Mellitus.

Detailed Description

The study enrolls 13 patients with Type 2 Diabetes in Zhongshan Hospital who are on treatment with insulin. After screening for the inclusion and exclusion criteria, eligible patients will receive insulin dosage titration set by the Insulin Titration System Based on Deep Learning in the intervention trial. The goal of insulin therapy was to achieve preprandial capillary blood glucose between 5.6-7.8 mmol/L and postprandial capillary glucose less than 10.0mmol/L. All patients are studied for 5 consecutive days or untill hospital discharge. For each patient, capillary glucose concentration was measured at 7 time points of fasting, after breakfast, before and after lunch, before and after dinner, and before bedtime a day using Glucometer (Glupad, Sinomedisite, China). Capillary glucose measurements were performed by the nurse staff according to standard procedures with a point-of-care testing device, which is integrated into the HIS system. And continuous glucose monitoring (CGM) was performed using flash glucose monitoring (Abbott Freestyle Libre, USA) placed on the upper left arm. This study will be conducted in the Department of Endocrinology, Zhongshan Hospital,Fudan University.

Conditions

Diabetes Mellitus Type 2 - Insulin-Treated

Study Design

• Allocation Method: NA
• Intervention Model: SINGLE_GROUP
• Primary Study Purpose: TREATMENT
• Blinding Strategy: NONE

Study Groups

AI

Insulin Titration System Based on Deep Learning

Group Type: EXPERIMENTAL

Insulin Titration System

Intervention Type: DEVICE

A noval insulin titration system, which is based on deep learning

Interventions

Insulin Titration System

A noval insulin titration system, which is based on deep learning

Intervention Type: DEVICE

Eligibility Criteria

Inclusion Criteria

• type 2 diabetes
• age of 18-75 years
• HbA1c between 7.0% and 11.0%.

Exclusion Criteria

• subjects with acute complications of diabetes, such as ketoacidosis or hyperglycemic hyperosmolar state;
• BMI ≥ 45kg/m2;
• women who are pregnant or breast-feeding;
• subjects with severe cardiac, hepatic, renal diseases; subjects with any psychiatric or psychological diseases;
• subjects with severe edema, infections or peripheral circulation disorders, receiving surgery during hospitalization;
• subjects who could not comply with the protocol

• Minimum Eligible Age: 18 Years
• Maximum Eligible Age: 75 Years
• Eligible Sex: ALL
• Accepts Healthy Volunteers: No

Sponsors

Shanghai Zhongshan Hospital

OTHER

Sponsor Role: lead

Responsible Party

Xiaoying Li

Professor

Responsibility Role: PRINCIPAL_INVESTIGATOR

Principal Investigators

Xiaoying Li

Role: PRINCIPAL_INVESTIGATOR

Shanghai Zhongshan Hospital

Locations

Department of Endocrinology, Zhongshan Hospital Fudan University

Shanghai, , China

Countries

China

References

Wang G, Liu X, Ying Z, Yang G, Chen Z, Liu Z, Zhang M, Yan H, Lu Y, Gao Y, Xue K, Li X, Chen Y. Optimized glycemic control of type 2 diabetes with reinforcement learning: a proof-of-concept trial. Nat Med. 2023 Oct;29(10):2633-2642. doi: 10.1038/s41591-023-02552-9. Epub 2023 Sep 14.

Reference Type: DERIVED

PMID: 37710000 (View on PubMed)

Other Identifiers

ZSE-202205

Identifier Type: -

Identifier Source: org_study_id