To address challenges in tongue image segmentation, such as discontinuous tongue edges and interference from complex backgrounds, this paper proposes a traditional Chinese medicine tongue image segmentation method based on Transformer feature channel fusion. First, multi-level feature maps containing both positional and feature information are obtained through a multi-stage convolution module. Next, an inverted pyramid network is introduced to match the dimensions of the multi-level feature maps. Finally, the skip connections of the traditional U-Net network are replaced with the CTrans module of UCTransNet to capture contextual information in the image better and achieve accurate segmentation of medical images. Dice coefficient and mean intersection-over-union (MIoU) are selected as evaluation criteria. Training, evaluation, and validation on a self-collected dataset of tongue images yield a Dice value of 96.81% and an MIoU value of 93.89%, indicating strong segmentation performance. The proposed method has a good segmentation effect on the tongue image dataset, and can accurately extract the features of the tongue body. This method can be used for standardized research in tongue diagnosis, improving the accuracy and reliability of tongue diagnosis. Additionally, it demonstrates strong generalization capabilities on other medical image datasets.
为了验证该方法的泛化能力,实验选取了医学开源数据集腺体分割数据集(Gland Segmentation,GlaS)[34]和多器官核分割数据集(Multiple organ Nuclear Segmentation,MoNuSeg)[35]评估本文所提的方法。腺体分割数据集是用于训练深度学习模型的一种数据集,旨在识别和分割腺体图像中的腺体细胞结构。该数据集包含来自不同组织类型的腺体图像,每个图像都有标记的腺体结构,共计165张图片。MoNuSeg数据集是一个多器官核分割数据集,由微观图像组成,该数据集包含7个器官的图像,涵盖了来自不同组织的多种细胞类型,包括细胞核、细胞膜、细胞质和细胞器等组织结构,共计44张图片。
MENGX Y, HUANGS Q. Origin and development of tongue diagnosis of TCM[J]. Liaoning Journal of Traditional Chinese Medicine, 2016, 43(5): 946-948. DOI: 10.13192/j.issn.1000-1719.2016.05.018 (Ch ).
WANGJ, CHANGP F, GUANJ, et al. Study on objectification research of tongue diagnosis in traditional Chinese medicine[J]. Journal of Basic Chinese Medicine, 2022, 28(9): 1531-1534. DOI: 10.19945/j.cnki.issn.1006-3250.2022.09.029 (Ch ).
ZHANGD, PANGW T, WANGK Y, et al. The present and future of the research on the objectification of TCM tongue diagnosis based on the microscopic angle[J]. World Science and Technology-Modernization of Traditional Chinese Medicine, 2022, 24(11): 4574-4579 .DOI:10.11842/wst.20211103006 (Ch ).
LID X, GUANJ, LIF. Development of tongue instrument and progress of its application in traditional Chinese medicine tongue characterization research[J]. World Chinese Medicine, 2017, 12(2): 456-460. DOI: 10.3969/j.issn.1673-7202.2017.02.054 (Ch ).
XIAOQ X, ZHENGX T, WUC C, et al. Research progress of tongue diagnosis in cirrhosis[J]. Medical Information, 2020, 33(19): 21-23. DOI: 10.3969/j.issn.1006-1959.2020.19.007 (Ch ).
SUH, ZHANGZ M, YONGW X, et al. General situation of objectification of TCM tongue diagnosis based on microecology of tongue coating[J]. Henan Traditional Chinese Medicine, 2021, 41(8): 1267-1270. DOI: 10.16367/j.issn.1003-5028.2021.08.0287 (Ch ).
SUNX, WANGD J, TIANC Y, et al. Objective and visual analysis of tongue diagnosis in traditional Chinese medicine[J]. Tianjin Journal of Traditional Chinese Medicine, 2021, 38(12): 1537-1543. DOI: 10.11656/j.issn.1672-1519.2021.12.10(Ch ).
QIZ, XUJ T, ZHANGZ F, et al. Progress in clinical application of tongue inspection objectivity based on digital image processing technique[J]. China Journal of Traditional Chinese Medicine and Pharmacy, 2015, 30(8): 2849-2851 (Ch).
ZHANGL Z, ZHOUW, ZHANGH L. Research progress of automatic tongue diagnosis technology[J]. Journal of Basic Chinese Medicine, 2023, 29(5): 871-876. DOI: 10.19945/j.cnki.issn.1006-3250.2023.05.035(Ch ).
JIANGS, HUJ, XIAC M, et al. A tongue image separation method based on Otsu threshold method and morphological adaptive correction[J]. Chinese High Technology Letters, 2017, 27(2): 150-155. DOI: 10.3772/j.issn.1002-0470.2017.02.006 (Ch ).
FUZ C, LIX Q, LIF F. Tongue image segmentation based on snake model and radial edge detection[J]. Journal of Image and Graphics, 2009, 14(4): 688-693 (Ch). DOI: 10.11834/jig.20090419 .
HUANGZ P, HUANGY S, YIF L, et al. An automatic tongue segmentation algorithm based on Otsu and region growing[J]. Lishizhen Medicine and Materia Medica Research, 2017, 28(12): 3062-3064. DOI: 10.3969/j.issn.1008-0805.2017.12.088 (Ch ).
YANGY P, QUY Y. General situation of research on cracked tongue[J]. Hunan Journal of Traditional Chinese Medicine, 2008, 24(6): 96-98. DOI: 10.16808/j.cnki.issn1003-7705.2008.06.064 (Ch ).
[29]
杨朝辉. 计算机舌诊中裂纹舌图像的诊断分类研究[D]. 哈尔滨: 哈尔滨工业大学, 2010.
[30]
YANGZ H. Research on diagnostic classification for fissured tongue images in computerized tongue diagnosis[D].Harbin: Harbin Institute of Technology, 2010(Ch).
LIM. Correlation study on the relationship between pathological tooth marked tongue and clinically common symptoms[J]. Journal of New Chinese Medicine, 2011, 43(11): 48-50. DOI: 10.13457/j.cnki.jncm.2011.11.024 (Ch ).
PANC M, ZHUP C, SISHIRS, et al. Study on the origin and clinical significance of toothed tongue[J]. Shaanxi Journal of Traditional Chinese Medicine, 2021, 42(9): 1267-1269. DOI: 10.3969/j.issn.1000-7369.2021.09.031 (Ch ).
[35]
RONNEBERGERO, FISCHERP, BROXT. U-net: Convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2015: 234-241. DOI:10.1007/978-3-319-24574-4_28 .
YANJ J, XUZ, GUOR, et al. Research on tongue image segmentation based on Mask R-CNN[J]. Modernization of Traditional Chinese Medicine and Materia Medica-World Science and Technology, 2020, 22(5):1532-1538 (Ch).
[40]
YANB, ZHANGS, YANGZ J, et al. Tongue segmentation and color classification using deep convolutional neural networks[J]. Mathematics, 2022, 10(22): 4286. DOI: 10.3390/math10224286 .
[41]
LIUW, CHENJ M, LIUB,et al. Tongue image segmentation and tongue color classification based on deep learning[J]. Digital Chinese Medicine,2022,5(3):253-263.DOI:10.1016/j.dcmed.2022.10.002 .
[42]
MATHEWJ K, SATHYALAKSHMIS. ExpACVO-Hybrid deep learning: Exponential anti corona virus optimization enabled hybrid deep learning for tongue image segmentation towards diabetes mellitus detection[J]. Biomedical Signal Processing and Control, 2023, 83: 104635. DOI: 10.1016/j.bspc.2023.104635 .
[43]
WANGH N, CAOP, WANGJ Q, et al. UCTransNet: Rethinking the skip connections in U-Net from a channel-wise perspective with transformer[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2022, 36(3): 2441-2449. DOI: 10.1609/aaai.v36i3.20144 .
[44]
LINT Y, DOLLÁRP, GIRSHICKR, et al. Feature pyramid networks for object detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). New York: IEEE Press, 2017: 936-944. DOI: 10.1109/CVPR.2017.106 .
[45]
DENGA P, LIUJ H, CHENQ Q, et al. Visual tracking with FPN based on transformer and response map enhancement[J]. Applied Sciences, 2022, 12(13): 6551. DOI: 10.3390/app12136551 .
LIUP, BIY X, ZHANGT Y, et al. Full-size object detection method optimized by attention mechanism[J]. Journal of Electronic Measurement and Instrumentation, 2023, 37(2): 193-203. DOI: 10.13382/j.jemi.B2205954 (Ch ).
[50]
HEJ Z, WANGJ, HANZ Y, et al. Cancer detection for small-size and ambiguous tumors based on semantic FPN and transformer[J]. PLoS One, 2023, 18(2): e0275194. DOI: 10.1371/journal.pone.0275194 .
SHANGGUANT J, DINGX M, WANGX H, et al. Multi-scale fetal image segmentation method based on attention Unet[J]. Control Engineering of China, 2023, 30(4): 722-729. DOI: 10.14107/j.cnki.kzgc.20220041(Ch ).
[53]
XUY, HOUS K, WANGX Y, et al. A medical image segmentation method based on improved UNet 3+ network[J]. Diagnostics, 2023, 13(3): 576. DOI: 10.3390/diagnostics13030576 .
[54]
HUANGX P, CHENJ X, CHENM Z, et al. TDD-UNet: Transformer with double decoder UNet for COVID-19 lesions segmentation[J]. Computers in Biology and Medicine, 2022, 151(Pt A): 106306. DOI: 10.1016/j.compbiomed.2022.106306 .
CHENJ, LIH Y, LANGX F, et al. Research on TCM image segmentation network model integrating Seg-UNet[J]. Modernization of Traditional Chinese Medicine and Materia Medica-World Science and Technology,2022,24(10):4073-4081.DOI:10.11842/wst.20220515001 (Ch ).
[57]
SIRINUKUNWATTANAK, PLUIMJ P W, CHENH, et al. Gland segmentation in colon histology images: The glas challenge contest[J]. Medical Image Analysis, 2017, 35: 489-502. DOI: 10.1016/j.media.2016.08.008 .
[58]
KUMARN, VERMAR, SHARMAS, et al. A dataset and a technique for generalized nuclear segmentation for computational pathology[J]. IEEE Transactions on Medical Imaging, 2017, 36(7): 1550-1560. DOI: 10.1109/TMI.2017.2677499 .
[59]
VALANARASUJ M J, OZA P, HACIHALILOGLUI, et al. Medical transformer: Gated axial-attention for medical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2021: 36-46.10.1007/978-3-030-87193-2_4. DOI: 10.1007/978-3-030-87193-2_4 .