Artificial Intelligence to Improve Cardiometabolic Risk Evaluation Using CT Scans

Artificial Intelligence to Improve Cardiometabolic Risk Evaluation Using CT Scans

To validate the ability of the FatHealth algorithm to identify individuals with pre-diabetes and with type 2 diabetes mellitus

This multicentre prospective study will evaluate the ability of the FatHealth technology to correctly identify individuals with pre-diabetes and diabetes, validating the technology against the current gold-standard diagnostic method, oral glucose tolerance testing.

Participants will be individuals who have undergone a CT scan of the chest (coronary CT angiogram [CCTA] or CT chest) as part of observational cohort studies.

Participants will be invited for an oral glucose tolerance test (OGTT), which is the current gold-standard method for detecting pre-diabetes and diabetes mellitus. All patients must have an evaluable OGTT. The study population will include:

1. Approximately 90 individuals who had a CCTA as part of their clinical care (45 identified as having an abnormal FatHealth algorithm calculation and 45 with a normal FatHealth algorithm calculation) will undergo OGTT which is evaluable; and
2. Approximately 90 individuals who had a chest CT as part of their clinical care (45 identified as having an abnormal FatHealth algorithm calculation and 45 with a normal FatHealth algorithm calculation) will undergo OGTT which is evaluable.

Inclusion Criteria:

• Participant is willing and able to give informed consent for participation in the study. Male or Female, aged 18 to 80 years.

• Body mass index (BMI) ≥ 25kg/m2

• FatHealth status assessed as the following:

  • Elevated FatHealth status (50% of participants)
  • Non-elevated FatHealth status (50% of participants)

Exclusion Criteria:

• Participant is unable or unwilling to give informed consent
• Participant is unable to understand English language
• Confirmed diagnosis of diabetes mellitus treated with oral medication or Insulin