1 January 2024 in Digital Health
Objective: Tongue segmentation as a basis for automated tongue recognition studies in Chinese medicine, which has defects such as network degradation and inability to obtain global features, which seriously affects the segmentation effect. This article proposes an improved model RTC_TongueNet based on DeepLabV3, which combines the improved residual structure and transformer and integrates the ECA (Efficient Channel Attention Module) attention mechanism of multiscale atrous convolution to improve the effect of tongue image segmentation. Methods: In this paper, we improve the backbone network based on DeepLabV3 by incorporating the transformer structure and an improved residual structure. The residual module is divided into two structures and uses different residual structures under different conditions to speed up the frequency of shallow information mapping to deep network, which can more effectively extract the underlying features of tongue image; introduces ECA attention mechanism after concat operation in ASPP (Atrous Spatial Pyramid Pooling) structure to strengthen information interaction and fusion, effectively extract local and global features, and enable the model to focus more on difficult-to-separate areas such as tongue edge, to obtain better segmentation effect. Results: The RTC_TongueNet network model was compared with FCN (Fully Convolutional Networks), UNet, LRASPP (Lite Reduced ASPP), and DeepLabV3 models on two datasets. On the two datasets, the MIOU (Mean Intersection over Union) and MPA (Mean Pixel Accuracy) values of the classic model DeepLabV3 were higher than those of FCN, UNet, and LRASPP models, and the performance was better. Compared with the DeepLabV3 model, the RTC_TongueNet network model increased MIOU value by 0.9% and MPA value by 0.3% on the first dataset; MIOU increased by 1.0% and MPA increased by 1.1% on the second dataset. RTC_TongueNet model performed best on both datasets. Conclusion: In this study, based on DeepLabV3, we apply the improved residual structure and transformer as a backbone to fully extract image features locally and globally. The ECA attention module is combined to enhance channel attention, strengthen useful information, and weaken the interference of useless information. RTC_TongueNet model can effectively segment tongue images. This study has practical application value and reference value for tongue image segmentation.
13 January 2025 in IEEE Access
Objective: Tongue image segmentation is a crucial step in the intelligent recognition of tongue diagnosis in Traditional Chinese Medicine (TCM). Existing deep learning-based tongue image segmentation models face issues such as poor versatility and insufficient expressiveness in zero-shot tasks. This study aims to construct an efficient model with zero-shot suitable for tongue image segmentation in TCM. Methods: We developed the Tongue-LiteSAM model by improving the SAM (Segment Anything Model) framework to suit tongue segmentation. Based on the basic SAM model, the improvement involved modifying the image encoder by integrating two lightweight ViT-Tiny image encoders, effectively reducing the model’s parameter count. Additionally, data perturbation techniques were employed to enhance the zero-shot segmentation capability of the model and ensure robust performance across different data sources. Results: Experiments conducted on six distinct tongue image datasets demonstrated that the Tongue-LiteSAM model outperformed traditional convolutional neural network-based models and transformers, the original SAM model, and other related improved models in tongue image segmentation tasks. Conclusion: The Tongue-LiteSAM model provides a more objective and consistent solution for tongue diagnosis, and has better zero-shot segmentation capabilities. By optimizing the model structure and data processing strategies, the accuracy and practicality of tongue diagnosis models are effectively improved, offering new technical support for the modernization and precision of TCM tongue diagnosis.