Skip to content

Validation of Machine Learning Based Personalized Nutrition Algorithm to Reduce Postprandial Glycemic Excursions Among North American Individuals With Newly Diagnosed Type 2 Diabetes

Validation of Machine Learning Based Personalized Nutrition Algorithm to Reduce Postprandial Glycemic Excursions Among North American Individuals With Newly Diagnosed Type 2 Diabetes

Status
Terminated
Phases
NA
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT03053518
Enrollment
22
Registered
2017-02-15
Start date
2017-06-30
Completion date
2018-01-31
Last updated
2020-09-21

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Type2 Diabetes

Keywords

Type 2 Diabetes, Hyperglycemia, Personal Nutrition Project

Brief summary

This is an initial validation study of the Personal Nutrition Project (PNP) algorithm in a North American population with recently diagnosed Type 2 Diabetes (T2D). This is a 2-stage, single-group feeding study in 20 individuals, including 10 participants managed with lifestyle alone, and 10 managed with lifestyle plus metformin.

Detailed description

The PNP algorithm, which uses a machine learning algorithm to predict postprandial glycemic, may be efficacious for generating tailored dietary advice to moderate the participant's glycemic response to food.

Interventions

A professional, blinded, continuous glucose monitoring device will be inserted on the back of the upper arm to measure interstitial glucose every 5 min for 4 times / day.

BEHAVIORALLifeStyle

Isocaloric diets (breakfast, lunch, dinner, and 2 snacks), which will be prepared and delivered daily, including 2 days each of low, moderate, and high glycemic load (GL) foods.

Sponsors

NYU Langone Health
Lead SponsorOTHER

Study design

Allocation
NON_RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
NONE

Intervention model description

Investigators will recruit approximately 5 cohorts of 4 participants each for the sample size of 20 participants. A new cohort will be randomized approximately every 4 weeks. The maximum number of study participants at any point in the study will be 10.

Eligibility

Sex/Gender
ALL
Age
21 Years to 70 Years
Healthy volunteers
No

Inclusion criteria

* Age \>21 years to \<70 years * Diagnosed with T2DM within 2 years with an HbA1c\<7% * Diabetes management by metformin or lifestyle intervention * Fasting C-peptide ≥ 0.5 mg/mL (0.17 nmol/L) (to exclude those for whom hyperglycemic exposure is driven by β-cell failure rather than dietary behaviors, as well as those requiring escalation of the medication regime) * Ownership a smart phone and are willing to use it to monitor multiple factors influencing glycemic response to glycemia (e.g., sleep, physical activity, diet, stress, medication, and hunger)

Exclusion criteria

* are unable or unwilling to provide informed consent; * are unable to participate meaningfully in an intervention that involves self-monitoring using software available in English (e.g., due to uncorrected sight impairment, illiterate, non-English-speaking, dementia); * are pregnant, are currently trying to become pregnant, or who become pregnant during the study * are institutionalized (e.g., in a nursing home or personal care facility, or those who are incarcerated and have limited control over self-management) * have had or are planning to have bariatric surgery during the study * have a history of heart disease, kidney disease, or retinopathy (to rule-out those with long-standing, undiagnosed T2D) * those with an active infection requiring antibiotics in the last 3 months or who develop an active infection requiring antibiotics during the study; * those who use acetaminophen and are unwilling or unable to discontinue its use during the study (acetaminophen affects CGM accuracy)39 * immunosuppressive drugs within three months prior to participation and * Chronically active inflammatory or neoplastic disease in the three years prior to enrollment. * Patients with known food allergy.

Design outcomes

Primary

MeasureTime frameDescription
Observed Incremental Area Under the Curve (iAUCobs)2 HoursObserved incremental area under the curve (iAUCobs) at 2 hours following each meal and snack will be evaluated via CGM using the Abbott Freestyle Libre Pro, which captures interstitial glucose every 5 minutes. A sensor is inserted into the participant's upper arm. Participants will be blinded to glycemia tracings.

Countries

United States

Participant flow

Participants by arm

ArmCount
Life Style + Metformin
Abbott Freestyle Libre Pro: A professional, blinded, continuous glucose monitoring device will be inserted on the back of the upper arm to measure interstitial glucose every 5 min for 4 times / day. LifeStyle: Isocaloric diets (breakfast, lunch, dinner, and 2 snacks), which will be prepared and delivered daily, including 2 days each of low, moderate, and high glycemic load (GL) foods.
22
Total22

Withdrawals & dropouts

PeriodReasonFG000
Overall StudyIneligible5
Overall StudyLost to Follow-up2
Overall StudyWithdrawal by Subject3

Baseline characteristics

CharacteristicLife Style + Metformin
Age, Continuous55 years
STANDARD_DEVIATION 10
Ethnicity (NIH/OMB)
Hispanic or Latino
1 Participants
Ethnicity (NIH/OMB)
Not Hispanic or Latino
21 Participants
Ethnicity (NIH/OMB)
Unknown or Not Reported
0 Participants
Race (NIH/OMB)
American Indian or Alaska Native
0 Participants
Race (NIH/OMB)
Asian
1 Participants
Race (NIH/OMB)
Black or African American
13 Participants
Race (NIH/OMB)
More than one race
0 Participants
Race (NIH/OMB)
Native Hawaiian or Other Pacific Islander
0 Participants
Race (NIH/OMB)
Unknown or Not Reported
1 Participants
Race (NIH/OMB)
White
7 Participants
Region of Enrollment
United States
22 participants
Sex: Female, Male
Female
13 Participants
Sex: Female, Male
Male
9 Participants

Adverse events

Event typeEG000
affected / at risk
deaths
Total, all-cause mortality
0 / 22
other
Total, other adverse events
0 / 22
serious
Total, serious adverse events
0 / 22

Outcome results

Primary

Observed Incremental Area Under the Curve (iAUCobs)

Observed incremental area under the curve (iAUCobs) at 2 hours following each meal and snack will be evaluated via CGM using the Abbott Freestyle Libre Pro, which captures interstitial glucose every 5 minutes. A sensor is inserted into the participant's upper arm. Participants will be blinded to glycemia tracings.

Time frame: 2 Hours

Population: There is not enough usable data collected to analyze primary outcome measure due to low adherence to the study.

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026