Using AI to Explore CPAP’s Impact on Abdominal Obesity in OSA, with Jennifer Prevot, MD

Obstructive sleep apnea (OSA) remains a prevalent and serious disorder closely linked with obesity, particularly abdominal adiposity. OSA is associated with an increased risk of cardiovascular disease (CVD), but the effectiveness of continuous positive airway pressure (CPAP) therapy, the standard treatment for OSA, in improving cardiovascular outcomes remains unclear.

Research presented at the American Thoracic Society (ATS) International Conference 2025 by Jennifer Prevot, MD, a pulmonary and critical care medicine fellow at Mount Sinai Hospital, addresses this gap in research by leveraging artificial intelligence to explore the relationship between OSA severity and abdominal obesity metrics, including visceral and subcutaneous adipose tissue (VAT and SAT) volumes and VAT metabolic activity, and to assess changes in these metrics following CPAP intervention using [18F]-Fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET) combined with magnetic resonance imaging (MRI).

“There is a lot of interest in obesity and obstructive sleep apnea, and CPAP has been the standard treatment for a very long time,” Prevot explained to HCPLive. “However, it's still questionable if it actually does anything in terms of cardiovascular risk factors. One thing that we know is that obesity kind of links everything, so I thought it would be interesting to look at abdominal obesity in particular and see the relationship it has with OSA and if CPAP therapy can modulate any risk factors.”

She and a group of investigators retrospectively analyzed PET/MRI scans from 115 adults with newly diagnosed OSA, both before and after 3 months of CPAP therapy. They determined OSA severity using portable sleep testing, defined by the respiratory disturbance index (pRDI). A deep learning model, using a transfer learning approach, segmented regions of interest (ROIs) within the subcutaneous and visceral adipose tissue compartments on MRI.

“Not only was the imaging technique that we used novel, because no one else has been using PET MRIs to look at visceral adipose tissue inflammation in the body, but we also established a relationship with our biomedical engineering department to automate the process in which we segment these compartments,” Prevot explained. “Not only do we have a novel imaging tool that we're using to look at metabolic activity of the abdominal tissue, but we are also applying a novel method to do this segmentation.”

Participants had an average age of 47.0 (standard deviation [SD], 11.97) years, were predominantly male (84.0%), and had an average BMI of 31.85 kg/m² (SD, 5.08). The mean pRDI was 32.53 events/hour (SD, 19.21).

Investigators noted OSA severity was not significantly associated with baseline VAT SUVmean (0.02%, confidence interval [CI], -0.21 to 0.25; P = .87) and CPAP therapy did not significantly alter VAT SUVmean (-2.17%, CI, -5.02 to 0.77; P = .15), VAT volume (-0.98%; CI, -2.41 to 0.47, P = .19), or weight (0.30%; CI, -0.83 to 1.43; P = .61). However, they called attention to a significant reduction in the VAT/SAT volume ratio (-1.70%; CI, -3.28 to 0.09; P = .04).

Looking ahead, Prevot emphasizes the potential of exploring these fat distribution metrics in the context of GLP-1 therapies, which are gaining attention for their metabolic benefits. She also highlights the possibility of evaluating changes in tongue fat, a known anatomical contributor to OSA, as an area for future investigation.

Editors’ note: Prevot has no relevant disclosures.

Reference

Prevot J, Patel H, Fauveau V, et al. Applying Artificial Intelligence to Measure Abdominal Adipose Metrics in Obstructive Sleep Apnea Before and After CPAP Therapy [abstract]. Am J Respir Crit Care Med 2025;211:A4759. https://doi.org/10.1164/ajrccm.2025.211.Abstracts.A4759