| Title: | Elucidate the correlation between features derived from abdominal MRI and the incidence and outcomes of metabolic diseases using multimodal data |
| Lead Institution: | Peking University People's Hospital. |
| Principal investigator: |
Application 145819
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
Many people around the world, more than a third of us, suffer from health problems called metabolic diseases like diabetes, obesity, and high blood pressure. These conditions can often lead to heart-related issues and death. Interestingly, certain aspects of our abdominal organs, like where fat is stored in abdominal organs, carries clues of these conditions. Our recent study found a link between the thickness of a part of the adrenal gland (found atop our kidneys) and conditions like obesity and high blood pressure. However, whether there is a causal relationship between the features of abdominal organs and metabolic diseases were unclear. Another big challenge in predicting who will get these metabolic conditions in the future. Old prediction methods, just based on general health info and age, aren't always accurate. However, newer methods using genetic info and cutting-edge technology called deep learning seem promising. Deep learning can process and analyze different types of data, from images (including abdominal MRI) to detailed genetic info, making predictions more precise. The key question is How do these abdominal MRI pictures connect to the start of metabolic health problems like diabetes? And, can we use different types of data to predict whether a person will develop these metabolic diseases? The UK Biobank (a big health data bank) has been useful in understanding these issues. Although previous researchers have used it to study many organs, they haven't looked at the adrenal gland in detail. There's a method called Mendelian randomization that uses genetic data to find out if there's a direct cause-and-effect relationship between two things. So, we're doing three main things: (1) Creating a new way to study the adrenal gland using MRI images. (2) Trying to understand the causal relationship between abdominal images on MRI and these metabolic health conditions. (3) Using deep learning to predict who might get these conditions in the future using various of measurements. We plan to finish this work within 3 years.1 Publication
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 17428 | Integrated Plasma Proteomics and Functional Analyses Reveal Hepatic CDHR2 as a Potential Therapeutic Target in MASLD | Yuanping Shi (+8) | 2026 | Diabetes Obesity and Metabolism |
Enabling scientific discoveries that improve human health