An interventional study of AI-CARE and Usual Care Group in Diabetes Mellitus, sponsored by Tan Tock Seng Hospital. Not yet recruiting at 1 site in Singapore. Open to participants aged 21 Years and older. Per ClinicalTrials.gov, last updated 2026-10-20.
Sponsored by Tan Tock Seng Hospital · Not applicable, Interventional, and Other
AI-CARE is an artificial intelligence-enabled clinical decision support mobile application designed to support prandial insulin dosing in individuals receiving MDI, incorporating a "human-in-the-loop" approach to ensure user verification of recommended doses. The application integrates AI-driven image-based carbohydrate estimation tailored to Asian dietary contexts, individualized patient-specific insulin parameters, and real-time CGM data to generate personalized insulin dose recommendations.
The primary objective of this study is to evaluate the feasibility, safety, and preliminary effectiveness of the AI-CARE mobile application among adults with diabetes receiving MDI. Feasibility will be assessed in terms of recruitment, retention, participant adherence, and application usability. Safety will be evaluated by monitoring the incidence of hypoglycaemia, hyperglycaemia, and other adverse events throughout the intervention period.
The study will also examine the preliminary effectiveness of AI-CARE on glycaemic and patient-reported outcomes compared with usual care. In addition, a qualitative process evaluation will be conducted to explore participants' experiences and perceptions of using AI-CARE, including perceived ease of use, usefulness, acceptability, and barriers and facilitators to real-world implementation. Findings from this phase will inform further refinement and optimization of the AI-CARE system and guide the design of a future large- scale randomized controlled trial.
Diabetes mellitus (DM) is a chronic metabolic disorder affecting an estimated 537 million adults worldwide, with prevalence projected to increase to 643 million by 2030. Poor glycaemic control is strongly associated with an increased risk of microvascular and macrovascular complications. Landmark trials, including the Diabetes Control and Complications Trial and the UK Prospective Diabetes Study, demonstrated that intensive glycaemic control achieved through multiple daily insulin injections (MDI) significantly reduces long-term complications. MDI typically consists of basal and bolus insulin therapy. Basal insulin is a long-acting insulin and is used to maintain background insulin levels and control glucose during fasting periods, including overnight and between meals. Bolus insulin, which is rapid-acting, is administered before meals to manage postprandial glucose excursions resulting from carbohydrate intake.
Clinical studies report that initiating MDI can reduce glycated haemoglobin (HbA1c) by approximately 1.5-2% within 3-4 months, with sustained reductions of up to 3-4% over 1 year. Epidemiological data indicate that each 1% reduction in HbA1c is associated with a 37% decrease in microvascular complications. Despite these benefits, a meta-analysis reported that only 23.75% of individuals with insulin-treated type 2 diabetes had HbA1c levels below 7%, while a cross-sectional study found that only 20% of individuals with type 1 diabetes attained the same target. Achieving optimal glycaemic control in individuals treated with MDI requires accurate pre- meal glucose assessment, precise carbohydrate estimation, and appropriate bolus insulin dose calculation. However, each of these components presents substantial real-world challenges.
Self-monitoring of blood glucose (SMBG) plays an important role in improving glycaemic control and empowering individuals with diabetes. It is particularly beneficial for those treated with insulin, as it helps determine bolus insulin dosing, facilitates insulin dose titration, and enables timely detection of hypoglycaemia. Despite strong recommendations for SMBG use among insulin-treated individuals, its uptake remains suboptimal in many Asian countries. In many Asian settings, blood glucose monitors and test strips are paid out of pocket, creating substantial financial barriers to regular testing. Furthermore, finger-prick testing is often perceived as painful and inconvenient. Patients may avoid monitoring due to frustration with high glucose readings, embarrassment when testing in public, fear of needles, workplace constraints, or limited motivation and self-efficacy. Collectively, these barriers result in inadequate pre-meal glucose monitoring, leading to inappropriate bolus dosing and suboptimal insulin titration, thereby contributing to poor glycaemic control despite insulin therapy.
Beyond glucose monitoring, carbohydrate counting-although a foundational component of MDI therapy-is cognitively demanding, time-consuming, and error-prone, particularly in multicultural dietary environments. In Singapore and other Asian settings, meals are commonly consumed in hawker centres, food courts, and home-based contexts where dishes are mixed, visually complex, and rarely accompanied by standardized nutritional information. Staple foods such as rice, noodles, congee, roti prata, and mixed-economy rice involve variable portion sizes and composite ingredients, making accurate carbohydrate estimation particularly challenging. Inaccurate estimation can lead to insulin underdosing or overdosing, resulting in postprandial hyperglycaemia or hypoglycaemia. Notably, mean carbohydrate-counting accuracy among individuals using MDI has been reported to be only 59%.
Even when glucose levels are measured and carbohydrate content estimated, bolus insulin calculation requires application of insulin-to-carbohydrate ratios (ICR), insulin sensitivity factors (ISF), and correction algorithms. This process demands numeracy and health literacy skills that are frequently limited among individuals with diabetes and are associated with poorer glycaemic outcomes. Difficulties in performing dose calculations may lead to reliance on fixed bolus doses, empirical estimation, incorrect administration, or omission of boluses. Indeed, 64% of patients have been shown to inaccurately estimate their prandial insulin requirements, and fixed-dose regimens have been associated with increased glycaemic variability and weight gain.
Collectively, these interrelated challenges-suboptimal glucose monitoring, inaccurate carbohydrate estimation, and numeracy burden in insulin dose calculation-contribute to postprandial glucose excursions and persistent hyperglycaemia despite insulin therapy. Addressing these barriers through integrated, automated clinical decision-support systems may therefore represent a promising strategy to enhance dosing accuracy, reduce cognitive burden, and improve overall glycaemic outcomes. AI-CARE is an artificial intelligence-enabled clinical decision support mobile application designed to support prandial insulin dosing in individuals receiving MDI, incorporating a "human-in-the-loop" approach to ensure user verification of recommended doses. The application integrates AI-driven image-based carbohydrate estimation tailored to Asian dietary contexts, individualized patient-specific insulin parameters, and real-time CGM data to generate personalized insulin dose recommendations.
The system comprises several key components. First, the image-based carbohydrate estimation module enables users to capture meal images at standardized angles (45-55° and 90°), after which the AI engine automatically estimates carbohydrate content. User-in- the-loop features, including portion adjustment and manual editing, are incorporated to enhance accuracy and allow verification. Second, real-time CGM data are integrated into the system, allowing current glucose values to be incorporated into the insulin dose calculation algorithm. Third, the application calculates personalized insulin dose recommendations based on parameters such as the ICR, ISF, and target glucose level, with predefined maximum bolus limits to enhance safety.
To ensure safe use, the application incorporates multiple safety features and feedback mechanisms, including dose caps, prompts to review glucose levels prior to insulin administration, and reminders to confirm dosing decisions. Although AI-CARE provides insulin dose recommendations, the final decision remains patient-driven, and users retain full autonomy to accept, modify, or override the suggested dose.
The primary objective of this study is to evaluate the feasibility, safety, and preliminary effectiveness of the AI-CARE mobile application among adults with diabetes receiving MDI. Feasibility will be assessed in terms of recruitment, retention, participant adherence, and application usability. Safety will be evaluated by monitoring the incidence of hypoglycaemia, hyperglycaemia, and other adverse events throughout the intervention period.
The study will also examine the preliminary effectiveness of AI-CARE on glycaemic and patient-reported outcomes compared with usual care. In addition, a qualitative process evaluation will be conducted to explore participants' experiences and perceptions of using AI-CARE, including perceived ease of use, usefulness, acceptability, and barriers and facilitators to real-world implementation. Findings from this phase will inform further refinement and optimization of the AI-CARE system and guide the design of a future large- scale randomized controlled trial.
This study adopts a mixed-methods design, comprising a 12-week feasibility pilot randomized controlled trial with an embedded qualitative process evaluation using focus group discussions.
11,089 studies on the registry are indexed under Diabetes Mellitus; 1,319 are open to participants now.
This study's planned enrollment of 48 is below the median of 80 across 8,491 interventional studies indexed under Diabetes Mellitus.
Browse Diabetes Mellitus studies →Tan Tock Seng Hospital is the lead sponsor of 108 studies on the registry; 19 are open to participants now.
Counted across the registry records on this site, refreshed daily.
Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.
Exclusion Criteria:
Participants randomized to the AI-CARE intervention arm will receive the intervention over a 12-week period, comprising five study contacts: a baseline visit (Visit 1), nurse-led telephone reviews at Week 1, 2 and 6, and a post-intervention visit (Visit 2) at Week 12. Following completion of the study, participants who provide consent will be invited to participate in semi-structured focus group discussions. Participants will be instructed to use the AI-CARE application prior to each main meal (typically two to three times daily) throughout the intervention period. Application usage data, including the frequency of meal image capture and acceptance of insulin dose recommendations, will be automatically recorded to assess intervention adherence and user engagement. A structured support infrastructure will be provided, including access to a study contact line during office hours and a predefined escalation pathway for urgent clinical concerns.
Device: AI-CARE
Participants in the usual care arm will continue to receive standard diabetes self-management practices without access to the AI-CARE application. Prandial insulin dosing will be calculated manually by participants using their individualized ICR and ISF, or according to instructions provided by their treating physician. All participants in this arm will wear a CGM sensor for study measurement purposes throughout the study period. Routine diabetes management and follow-up will continue in accordance with standard outpatient care practices. The study visit schedule includes two in-person visits: Visit 1 at baseline for enrolment, baseline assessments, and CGM sensor application, structured nurse-led telephone reviews at Week 1, 2 and 6, and Visit 2 at Week 12 for post-intervention outcome assessment. Following completion of the study, participants who provided consent will be invited to participate in semi-structured focus group discussions.
Other: Usual Care Group
AI-CARE is an artificial intelligence-enabled clinical decision support mobile application designed to support prandial insulin dosing in individuals receiving MDI, incorporating a "human-in-the-loop" approach to ensure user verification of recommended doses. The application integrates AI-driven image-based carbohydrate estimation tailored to Asian dietary contexts, individualized patient-specific insulin parameters, and real-time CGM data to generate personalized insulin dose recommendations.
Participants randomized to the usual care arm will continue to receive standard diabetes self-management practices without access to the AI-CARE application. Usual care includes manual carbohydrate counting using conventional methods, such as estimating carbohydrate content based on food labels, dietary guides, or prior diabetes education. Prandial insulin dosing will be calculated manually by participants using their individualized ICR and ISF, or according to instructions provided by their treating physician. All participants in this arm will wear a CGM sensor for study measurement purposes throughout the study period. Routine diabetes management and follow-up will continue in accordance with standard outpatient care practices. The study visit schedule for the usual care group includes two in-person visits: Visit 1 at baseline for enrolment, baseline assessments, and CGM sensor application, and Visit 2 at Week 12 for post-intervention outcome assessment.
Time to achieve target sample size
Recruitment feasibility will be evaluated by measuring the time (in weeks) required to achieve the target sample size.
Time frame: From enrollment to the end of follow-up at 12 weeks
Eligibility rate
Recruitment feasibility will also be measured by calculating the eligibility rate, which is the proportion of eligible participants among those screened.
Time frame: From enrollment to the end of follow-up at 12 weeks
Recruitment rate
Recruitment feasibility will also be measured by calculating the recruitment rate, which is the proportion of enrolled participants among those eligible.
Time frame: From enrollment to the end of follow-up at 12 weeks
Retention and attrition rates
Retention and attrition rates will also be calculated, with attrition defined as the proportion of participants who withdraw or are lost to follow-up during the 12-week intervention period.
Time frame: From enrollment to the end of follow-up at 12 weeks
Intervention engagement
Intervention engagement will be assessed through application usage metrics, including the frequency of daily insulin calculator activation and the proportion of meals supported by AI-CARE. Acceptability will be further inferred from app usage patterns.
Time frame: From enrollment to the end of follow-up at 12 weeks
Intervention usability
Usability will be evaluated using the validated mHealth App Usability Questionnaire (MAUQ) for Standalone mHealth Apps Used by Patients, which will be administered at Week 12. In the MAUQ questionnaire, there are 18 statements, for which participants indicate their level of agreement on a 7-point scale (1 - strongly disagree, 2 - disagree, 3 - somewhat disagree, 4 - neither agree nor disagree, 5 - somewhat agree, 6 - agree, 7 - strongly agree). To determine the usability of an app, calculate the total and determine the average of the responses to all statements. The higher the overall average, the higher the usability of the app. There is also a section for participants' written comments. Acceptability will be further inferred from participant feedback.
Time frame: From enrollment to the end of follow-up at 12 weeks
Changes in glycaemic control (HbA1c)
Changes in glycaemic control will be assessed using HbA1c (%).
Time frame: From enrollment to the end of follow-up at 12 weeks
Continuous glucose monitoring-derived metrics
Continuous glucose monitoring-derived metrics include Time in Range (TIR), Time Above Range (TAR) and Time Below Range (TBR). TIR is the percentage of time that blood glucose stays within the normal range. TAR is the percentage of time that blood glucose rises above the target range. TBR is the percentage of time that blood glucose falls below the target range.
Time frame: From enrollment to the end of follow-up at 12 weeks
Body Weight
Body weight will be measured in kilograms and evaluated to explore broader metabolic effects of the intervention.
Time frame: From enrollment to the end of follow-up at 12 weeks
Body Mass Index (BMI)
Weight (in kilograms) and height (in meters) will be aggregated to report BMI in kg/m\^2) and evaluated to explore broader metabolic effects of the intervention.
Time frame: From enrollment to the end of follow-up at 12 weeks
Blood Pressure
Blood pressure will be reported in mmHg and evaluated to explore broader metabolic effects of the intervention.
Time frame: From enrollment to the end of follow-up at 12 weeks
Low-Density Lipoprotein Cholesterol
Low Density Lipoprotein (LDL) reported in mmol/L will also be evaluated to explore broader metabolic effects of the intervention.
Time frame: From enrollment to the end of follow-up at 12 weeks
Treatment satisfaction
Patient-reported outcomes will assess changes in diabetes treatment satisfaction as measured using the Diabetes Treatment Satisfaction Questionnaire (DTSQ). The 8 questions in the DTSQ are concerned with the treatment for participants' diabetes (including insulin, tablets and/or diet) and their experience over the past few weeks. The DTSQ (status) questionnaire will be administered at baseline, with each question answered on a 7-point scale (0: very dissatisfied; 6: very satisfied). The maximum score is 48, minimum score is 0. The DTSQ (change) questionnaire will be administered at week 12, with the 7-point scale indicating instead the extent to which participants have experienced changes in the 8 questions (-3: much less satisfied now; 3: much more satisfied now). The maximum score is 24, minimum score is -24.
Time frame: From enrollment to the end of follow-up at 12 weeks
Quality of Life
Changes in quality of life will be assessed using the Audit of Diabetes-Dependent Quality of Life (ADDQoL) questionnaire. The ADDQoL consists of two overview items; one measures generic overall QoL and a further 19 items are concerned with the impact of diabetes on specific aspects of life. The scales range from -3 to +1 for 19 life domains (impact rating) and from 0 to +3 in attributed importance (importance rating). A weighted score for each domain is calculated as a multiplier of impact rating and importance rating (ranging from -9 to +3). Lower scores reflect poorer QoL. Finally, a mean weighted impact score (ADDQOL score) is calculated for the entire scale across all applicable domains.
Time frame: From enrollment to the end of follow-up at 12 weeks
Distress
Changes in diabetes-related distress will be measured using the Diabetes Distress Scale (DDS17). The DDS17 lists 17 potential problem areas that people with diabetes may experience within the past month. Participants indicate on a 6-point scale the extent to which each of the 17 items has caused distress to them (1: not a problem, 6: a very serious problem). This yields a total diabetes distress score pls 4 subscale scores, addressing different kinds of distress; Emotional burden, physician distress, regimen distress and interpersonal distress. To score, sum up the participant's responses and divide by the number of items in that scale. Current research suggests that a mean item score 2.0 - 2.9 should be considered 'moderate distress,' and a mean item score \> 3.0 should be considered 'high distress.'
Time frame: From enrollment to the end of follow-up at 12 weeks
Intervention Safety
Safety will be monitored throughout the study, with adverse events including severe hypoglycaemia (requiring third-party assistance), diabetic ketoacidosis, and clinically significant hypoglycaemia identified through continuous glucose monitoring (CGM) data review. Any serious or potentially intervention-related adverse event will prompt immediate clinical review and, if necessary, consideration of protocol modification
Time frame: From enrollment to the end of follow-up at 12 weeks
Documents are hosted by the registry — open the source record to download them.
Plan to share: No
Start is estimated for Oct 2026; primary completion for Oct 2028, 24 months later.
Of 2,196 non-industry interventional studies with no phase under Diabetes Mellitus, half reached primary completion within 21 months of starting (the median), counting those still running; a quarter within 10 and three quarters within 38. This study's planned 24 months is above that.
Counting every sponsor, half of 2,388 studies reached it within 20 months; a quarter within 9 and three quarters within 37.
How these figures are worked out, and which studies are left out of them.
From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗
This study is not yet recruiting, as verified in Oct 2026. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.