The location of where fat is stored in the body is the main determinant of whether a person develops type 2 diabetes, with fat under the skin being less harmful than fat in deeper areas of the abdomen.

Researchers at the German Center for Diabetes Research (DZD) say that whole-body MR imaging can help visualize the distribution of fat in the body and indicate who is at risk early for developing diabetes. To prove their point, the scientists conducted a study that combined whole-body MR with deep-learning algorithms.

"An analysis of the model results showed that fat accumulation in the lower abdomen plays a crucial role in diabetes detection. Further, additional analysis also showed that a proportion of people with prediabetes, as well as people with a diabetes subtype that can lead to kidney disease, can also be identified via MR scans,” said the author, professor Robert Wagner, in a statement.
Wagner and his colleagues applied deep-learning networks to whole-body MR scans of 2,000 people who were tested with oral glucose tolerance tests (OGTT), which are used to screen for impaired glucose metabolism and diagnose diabetes. It was the OGTT tests that served as training material for the deep-learning networks.

The researchers are now using the findings to investigate the biological regulation of body fat distribution so they can identify the causes for diabetes through new methods. This includes the use of AI, with the goal being to find preventive and therapeutic approaches for addressing type 2 diabetes.

Interest in whole-body MR continues to grow among medical imaging providers. A recent study out of King’s College London found that whole-body MR (WBMR) may be better than PET/CT with 18F-fluorodeoxyglucose (FDG) at detecting more myeloma-defining disease, a bone marrow cancer that 140,000 new patients worldwide are diagnosed with each year and less than 50% of all patients survive after five years.

While CT can show myeloma-associated bone destruction, diagnoses are often in the late stages with this modality. WBMR was found to be more sensitive than PET/CT or CT alone for bone marrow infiltration by myeloma, and helped clinicians treat 7% more patients than PET/CT did. It also changed how patients would have been managed by their providers in 24% of cases by allowing them to get treatment earlier, as it is more sensitive to a lower skeletal disease burden. As a result, patients assessed with WBMR showed improved outcomes.

DZD is one of six German Centers for Health Research and is part of Tübingen University Hospital.

The findings of the DZD study were published in JCI Insight.

MRI Homepage