Advanced CT scan analysis may help identify different forms of sarcoidosis: Study
Researchers have identified a new method for analyzing chest CT scans that may help physicians better understand the different forms of sarcoidosis, a complex inflammatory lung disease that affects more than 150,000 people in the United States. In a new study published in Scientific Reports, scientists at National Jewish Health and colleagues used a computer-based imaging technique known as radiomics to analyze subtle patterns in lung scans from 320 people with sarcoidosis. The approach identified four distinct imaging profiles that were linked to differences in lung function and disease severity and gave insight into potential uses of radiomics to advance diagnosis and patient care.
Sarcoidosis most commonly affects the lungs and can lead to inflammation, scarring and breathing difficulties. Doctors typically assess lung involvement by visually reviewing imaging scans, but interpretations can vary among specialists.
Radiomics offers a more objective approach by using advanced algorithms to measure hundreds of quantitative features from medical images, capturing subtle patterns that may not be readily visible to the human eye.
“We found that radiomic analysis of CT scans can reveal distinct patterns of lung abnormalities in sarcoidosis,” said Tasha Fingerlin, PhD, vice chair of the Department of Immunology and Genomic Medicine at National Jewish Health and co-senior author of the study. “These patterns were associated with differences in lung function, suggesting that this approach may help us better understand how the disease varies from patient to patient.”
“Using radiomic analysis could allow providers to evaluate patients’ pulmonary status and radiographs in an automated way,” said Lisa Maier, MD, chief of the Division of Environmental and Occupational Health Sciences, head of the World Association of Sarcoidosis and Granulomatous Disease (WASOG) Sarcoidosis Center of Excellence at National Jewish Health and co-senior author of the study.
The research team analyzed previous high-resolution CT scans from participants in the Genomic Research in Alpha-1 Antitrypsin Deficiency and Sarcoidosis (GRADS) Study, one of the largest and most comprehensive sarcoidosis research cohorts in the United States. Using machine-learning clustering techniques, the investigators identified four patient groups based on their features seen in the images. Some patients showed relatively few abnormalities, while others showed patterns consistent with more extensive inflammation or fibrosis. Importantly, the radiomic groups accounted for differences in lung function even after adjusting for traditional imaging assessments used in clinical practice.
“Current staging systems are helpful, but they don’t always capture the full complexity of what we see in the lungs of people with sarcoidosis,” Dr. Fingerlin said. “Radiomics allows us to quantify those patterns in a more detailed and reproducible way.”
Because the radiomics analysis can be performed quickly and automatically using open-source software, the researchers say it could eventually help clinicians analyze large numbers of scans and better track disease patterns over time.
“Radiomics has the potential to complement the expertise of radiologists by providing objective measurements of lung abnormalities, identifying disease subtypes, monitoring progression and potentially guiding more personalized treatment strategies,” said Dr. Fingerlin.
The researchers note that further studies are needed to determine how radiomic analysis could be used in routine clinical care.
“There is promise for significant impact on patient care, especially in regions where there is no expert in sarcoidosis radiology, which is much of the country and certainly most areas in the Far West,” said Dr. Maier. “Radiomics could also expedite care in clinics with rapid turnaround for patients at specialized centers and revolutionize the way we interpret CT scans for research and clinical trials.”
Reference:
Carlson, N.E., Lippitt, W.L., Ryan, S.M. et al. Radiomic profiling of chest CT in a cohort of sarcoidosis cases. Sci Rep 16, 9695 (2026). https://doi.org/10.1038/s41598-026-39384-9
