College of Information Science and Technology, Zhejiang Shuren University, Hangzhou, 310015, China.
Zhejiang-Netherlands Joint Laboratory for Digital Diagnosis and Treatment of Oral Diseases, Zhejiang Shuren University, Hangzhou, 310015, China.
Sci Rep. 2024 Mar 1;14(1):5068. doi: 10.1038/s41598-024-55522-7.
Using deep learning technology to segment oral CBCT images for clinical diagnosis and treatment is one of the important research directions in the field of clinical dentistry. However, the blurred contour and the scale difference limit the segmentation accuracy of the crown edge and the root part of the current methods, making these regions become difficult-to-segment samples in the oral CBCT segmentation task. Aiming at the above problems, this work proposed a Difficult-to-Segment Focus Network (DSFNet) for segmenting oral CBCT images. The network utilizes a Feature Capturing Module (FCM) to efficiently capture local and long-range features, enhancing the feature extraction performance. Additionally, a Multi-Scale Feature Fusion Module (MFFM) is employed to merge multiscale feature information. To further improve the loss ratio for difficult-to-segment samples, a hybrid loss function is proposed, combining Focal Loss and Dice Loss. By utilizing the hybrid loss function, DSFNet achieves 91.85% Dice Similarity Coefficient (DSC) and 0.216 mm Average Symmetric Surface Distance (ASSD) performance in oral CBCT segmentation tasks. Experimental results show that the proposed method is superior to current dental CBCT image segmentation techniques and has real-world applicability.
使用深度学习技术对口腔 CBCT 图像进行分割，以便于临床诊断和治疗，是临床牙科领域的重要研究方向之一。然而，目前方法的冠状边缘和根部的轮廓模糊和尺度差异限制了分割精度，使得这些区域成为口腔 CBCT 分割任务中的难分割样本。针对上述问题，本研究提出了一种用于口腔 CBCT 图像分割的难分割焦点网络（DSFNet）。该网络利用特征捕获模块（FCM）有效地捕获局部和长程特征，增强了特征提取性能。此外，还采用了多尺度特征融合模块（MFFM）来融合多尺度特征信息。为了进一步提高难分割样本的损失比，提出了一种混合损失函数，结合了焦点损失和 Dice 损失。通过利用混合损失函数，DSFNet 在口腔 CBCT 分割任务中实现了 91.85%的 Dice 相似系数（DSC）和 0.216mm 的平均对称表面距离（ASSD）性能。实验结果表明，该方法优于现有的牙科 CBCT 图像分割技术，具有实际应用价值。
