Parathyroid glands in primary hyperparathyroidism exhibit distinct near-infrared autofluorescence (NIRAF) signatures, with normal glands showing higher intensity (2.68 vs 2.09 pixels, p<.0001) and lower heterogeneity (0.11 vs 0.18, p<.0001). A deep learning model developed with these parameters achieved 83.3% precision and recall, and an area under the precision-recall curve of 0.908. These findings suggest that NIRAF imaging and machine learning can significantly assist in differentiating normal from diseased parathyroid glands during surgery.
• Why it matters: Improves differentiation of diseased and normal parathyroid glands intraoperatively.
Journal Article by Akgun E, Ibrahimli A and Berber E in J Am Coll Surg
Copyright © 2024 by the American College of Surgeons. Published by Wolters Kluwer Health, Inc. All rights reserved.