Primary study · included by 1 review
Automatic 3-dimensional quantification of orthodontically induced root resorption in cone-beam computed tomography images based on deep learning.
Zheng 2025 · Zheng Qianhan, Ma Lei, Wu Yongjia, et al. · American journal of orthodontics and dentofacial orthopedics : official publication of the American Association of Orthodontists, its constituent societies, and the American Board of Orthodontics · 2025
Included by
| Outcome this study informs | Pooled effect | Evidence | Certainty |
|---|---|---|---|
Pooled sensitivity for detection of OIRR AI models (CNN or other deep learning) applied to CBCT for detection and quantification of OIRR vs Manual CBCT interpretation or manual measurement/segmentation (reference standard) | Sensitivity 0.903 [0.818, 0.989] | 3 studies | Moderate |
Pooled sensitivity for detection of OIRR AI models (CNN or other deep learning) applied to CBCT for detection and quantification of OIRR vs Manual CBCT interpretation or manual measurement/segmentation (reference standard) | Sensitivity 0.903 [0.818, 0.989] | 3 studies | GRADE not reported |
Pooled sensitivity for detection of OIRR AI models (CNN or other deep learning) applied to CBCT for detection and quantification of OIRR vs Manual CBCT interpretation or manual measurement/segmentation (reference standard) | Sensitivity 0.93 [0.89, 0.96] | 5 studies · 420 participants | Moderate |
Agreement of AI with manual quantification of OIRR (ICC) AI models (CNN or other deep learning) applied to CBCT for detection and quantification of OIRR vs Manual CBCT interpretation or manual measurement/segmentation (reference standard) | ICC 1.000 | 3 studies · 160 participants | High |
Discriminative accuracy for OIRR detection (AUC) AI models (CNN or other deep learning) applied to CBCT for detection and quantification of OIRR vs Manual CBCT interpretation or manual measurement/segmentation (reference standard) | AUC 0.955 | 2 studies | Low |
Also linked to PM41744917-P2-O1, which does not match a current question in this review.
1 including review in total.