Deep learning improves detection of changes in chronic sinus disease on CT scans
A deep-learning tool developed to analyze sinus CT scans may provide a more sensitive way to detect treatment-related changes in patients with chronic rhinosinusitis with nasal polyps, according to new research led in part by researchers at National Jewish Health.
Researchers from National Jewish Health have developed a deep-learning tool that may enhance the detection of treatment-related changes in patients with chronic rhinosinusitis with nasal polyps through CT scans. The study, published in the International Forum of Allergy & Rhinology, compared the automated deep-learning-based sinus severity score (SSS) to the traditional Lund-Mackay score, a manual method used by clinicians to evaluate sinus disease visually on CT scans.
The automated method proved to be more sensitive to changes following treatment, offering a more precise way to measure treatment response. Chronic rhinosinusitis with nasal polyps is a persistent inflammatory condition that can block nasal passages and sinuses, making CT imaging crucial for assessing disease extent and evaluating new therapies.
However, the widely used Lund-Mackay scoring system relies on radiologists' subjective visual assessments, which may not capture smaller changes over time. The study's findings, limited to two clinical trials of one treatment, highlight the potential of deep learning to improve the objectivity and precision of CT imaging measurements, making it a more useful tool for evaluating therapies and understanding sinus disease response.
Written by urgent.news from Medical Xpress's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
