Quantitative Ultrasound Tracks Induced Gum Inflammation in Pigs Across Five Timepoints
Preprint demonstrates QUS feasibility for longitudinal periodontal inflammation tracking in a pig model. ACS and BSI changed reliably with induced disease; classification accuracy reached 92 percent at one site. Small sample size and absence of human data remain key constraints.
Researchers induced inflammation at three interproximal sites per quadrant using complementary methods and acquired QUS data longitudinally. Both parameters changed significantly by week 2 or 4, with backscatter intensity rising and attenuation slope falling; sex and site stratification showed consistent BSI effects but more variable ACS results. The 2D classifier separated baseline from week-2 scans at 92 percent, 82 percent, and 74 percent accuracy for the three sites tested. This staggered pig design provides within-animal controls yet remains limited by small N and lack of histologic ground truth at every timepoint. Translation to humans will require validation against probing depths and radiographs in a powered clinical cohort. If thresholds hold, chairside QUS could replace serial radiographs for monitoring early periodontal changes within five years.
HELIX: Human pilot trials of chairside QUS periodontal monitoring will begin within 24 months once a 50-patient validation study confirms ACS and BSI thresholds against clinical attachment loss.
Sources (2)
- [1]Primary Source(https://arxiv.org/abs/2607.05676)
- [2]Supporting Source(https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6314465/)