Department of Orthodontics, Faculty of Dentistry, Eskisehir Osmangazi University, Eskişehir, Turkey.
Department of Orthodontics, Faculty of Dentistry, Recep Tayyip Erdoğan University, Rize, Turkey.
BMC Med Imaging. 2024 Oct 30;24(1):294. doi: 10.1186/s12880-024-01478-z.
We explored whether the feature aggregation and refinement network (FARNet) algorithm accurately identified posteroanterior (PA) cephalometric landmarks.
We identified 47 landmarks on 1,431 PA cephalograms of which 1,177 were used for training, 117 for validation, and 137 for testing. A FARNet-based artificial intelligence (AI) algorithm automatically detected the landmarks. Model effectiveness was calculated by deriving the mean radial error (MRE) and the successful detection rates (SDRs) within 2, 2.5, 3, and 4 mm. The Mann-Whitney U test was performed on the Euclidean differences between repeated manual identifications and AI trials. The direction in differences was analyzed, and whether differences moved in the same or opposite directions relative to ground truth on both the x and y-axis.
The AI system (web-based CranioCatch annotation software (Eskişehir, Turkey)) identified 47 anatomical landmarks in PA cephalograms. The right gonion SDRs were the highest, thus 96.4, 97.8, 100, and 100% within 2, 2.5, 3, and 4 mm, respectively. The right gonion MRE was 0.94 ± 0.53 mm. The right condylon SDRs were the lowest, thus 32.8, 45.3, 54.0, and 67.9% within the same thresholds. The right condylon MRE was 3.31 ± 2.25 mm. The AI model's reliability and accuracy were similar to a human expert's. AI was better at four skeleton points than the expert, whereas the expert was better at one skeletal and seven dental points (P < 0.05). Most of the points exhibited significant deviations along the y-axis. Compared to ground truth, most of the points in AI and the second trial showed opposite movement on the x-axis and the same on the y-axis.
The FARNet algorithm streamlined orthodontic diagnosis.
我们探讨了特征聚合和细化网络（FARNet）算法是否能准确识别前后位（PA）头影测量标志点。
我们在 1431 张 PA 头影片中识别出 47 个标志点，其中 1177 个用于训练，117 个用于验证，137 个用于测试。基于 FARNet 的人工智能（AI）算法自动检测标志点。通过计算 2、2.5、3 和 4 mm 内的平均放射状误差（MRE）和成功检测率（SDR）来评估模型的有效性。对重复手动识别和 AI 试验之间的欧几里得差异进行曼-惠特尼 U 检验。分析差异的方向，以及差异在 x 和 y 轴上相对于真实值是朝同一方向还是相反方向移动。
AI 系统（基于网络的 CranioCatch 注释软件（土耳其埃斯基谢希尔））在 PA 头影片中识别出 47 个解剖标志点。右侧髁突的 SDR 最高，分别为 96.4、97.8、100 和 100%，在 2、2.5、3 和 4 mm 内。右侧髁突的 MRE 为 0.94±0.53 mm。右侧髁突的 SDR 最低，分别为 32.8、45.3、54.0 和 67.9%，在相同阈值内。右侧髁突的 MRE 为 3.31±2.25 mm。AI 模型的可靠性和准确性与人类专家相似。AI 在四个骨骼点上优于专家，而专家在一个骨骼点和七个牙齿点上优于 AI（P<0.05）。大多数点在 y 轴上表现出显著的偏差。与真实值相比，AI 和第二次试验中的大多数点在 x 轴上表现出相反的运动，而在 y 轴上则表现出相同的运动。
FARNet 算法简化了正畸诊断。
