- HIGH
- Research
- Tier 1
- Peer-Reviewed Research
AI enabled automated digital image analysis for peri-implant health monitoring
- Source
- Clinical Oral Implants Research
- Author
- Xinyu Wu, Xindi Wei, Xiaolei Lv, Xiaojiao Fu, Ziqi He, Wanting Yin, Hongchang Lai, Junyu Shi
- Originally at
- onlinelibrary.wiley.com(opens in a new tab)
Summary and scoring by The Bell Brief (Dr. Jennifer Bell) using the Drill-Down Protocol. Not the original publisher.
Why it matters for dental
Implantologists and periodontists gain an objective, non-invasive metric that correlates with bleeding on probing, potentially reducing subjective visual assessment of peri-implant erythema.
Key points
- 01
U-Net model segmented peri-implant mucosa on intraoral scans with Dice coefficient 0.69 ± 0.11 and ICC > 0.95 for colorimetric calculation.
- 02
Digital Erythema Score identified mucosal erythema with AUC 0.9228 internally and 0.9018 on an external 30-implant set.
- 03
YOLO model localized inflammatory regions at mAP 0.8968 internally and 0.6820 externally.
- 04
DES framework improved diagnostic performance and inter-observer consistency for clinicians with varying experience levels in the external validation.
Who should care
- Specialist
- Academia / research
Read the original
Clinical Oral Implants Research
Full reporting and any paywalled content live on onlinelibrary.wiley.com. We summarize and score; we do not republish.