基于深度学习技术的侧位头颅定位自动化检测评估。

Evaluation of automated detection of head position on lateral cephalometric radiographs based on deep learning techniques.

机构信息

State Key Laboratory of Oral Diseases & National Clinical Research Center for Oral Diseases, West China Hospital of Stomatology, Sichuan University, Chengdu 610041, China.

Chongqing University Three Gorges Hospital, Chongqing 404031, China.

出版信息

Ann Anat. 2023 Oct;250:152114. doi: 10.1016/j.aanat.2023.152114. Epub 2023 Jun 9.

DOI:10.1016/j.aanat.2023.152114

PMID:37302431

Abstract

BACKGROUND

Lateral cephalometric radiograph (LCR) is crucial to diagnosis and treatment planning of maxillofacial diseases, but inappropriate head position, which reduces the accuracy of cephalometric measurements, can be challenging to detect for clinicians. This non-interventional retrospective study aims to develop two deep learning (DL) systems to efficiently, accurately, and instantly detect the head position on LCRs.

METHODS

LCRs from 13 centers were reviewed and a total of 3000 radiographs were collected and divided into 2400 cases (80.0 %) in the training set and 600 cases (20.0 %) in the validation set. Another 300 cases were selected independently as the test set. All the images were evaluated and landmarked by two board-certified orthodontists as references. The head position of the LCR was classified by the angle between the Frankfort Horizontal (FH) plane and the true horizontal (HOR) plane, and a value within - 3°- 3° was considered normal. The YOLOv3 model based on the traditional fixed-point method and the modified ResNet50 model featuring a non-linear mapping residual network were constructed and evaluated. Heatmap was generated to visualize the performances.

RESULTS

The modified ResNet50 model showed a superior classification accuracy of 96.0 %, higher than 93.5 % of the YOLOv3 model. The sensitivity&recall and specificity of the modified ResNet50 model were 0.959, 0.969, and those of the YOLOv3 model were 0.846, 0.916. The area under the curve (AUC) values of the modified ResNet50 and the YOLOv3 model were 0.985 ± 0.04 and 0.942 ± 0.042, respectively. Saliency maps demonstrated that the modified ResNet50 model considered the alignment of cervical vertebras, not just the periorbital and perinasal areas, as the YOLOv3 model did.

CONCLUSIONS

The modified ResNet50 model outperformed the YOLOv3 model in classifying head position on LCRs and showed promising potential in facilitating making accurate diagnoses and optimal treatment plans.

摘要

背景

侧颅面 X 光片（LCR）对颌面疾病的诊断和治疗计划至关重要，但临床医生很难发现因头部位置不当而导致的头测测量准确性降低的问题。这项非介入性回顾性研究旨在开发两个深度学习（DL）系统，以高效、准确和即时地检测 LCR 上的头部位置。

方法

回顾了 13 个中心的 LCR，共收集了 3000 张 X 光片，分为 2400 例（80.0%）在训练集中和 600 例（20.0%）在验证集中。另外 300 例被独立选择作为测试集。所有图像均由两名经过董事会认证的正畸医生进行评估和标记作为参考。LCR 的头部位置通过法兰克福水平面（FH）与真实水平面（HOR）之间的角度来分类，在-3°至 3°范围内的被认为是正常的。构建并评估了基于传统定点法的 YOLOv3 模型和具有非线性映射残差网络的改进 ResNet50 模型。生成热力图以可视化性能。

结果

改进的 ResNet50 模型的分类准确率为 96.0%，高于 YOLOv3 模型的 93.5%。改进的 ResNet50 模型的灵敏度&召回率和特异性分别为 0.959、0.969，而 YOLOv3 模型的灵敏度&召回率和特异性分别为 0.846、0.916。改进的 ResNet50 模型和 YOLOv3 模型的曲线下面积（AUC）值分别为 0.985±0.04 和 0.942±0.042。显著图表明，改进的 ResNet50 模型不仅考虑了眶周和鼻周区域，还考虑了颈椎的对齐方式，而 YOLOv3 模型则没有。