JoVE Journal > Medicine
Summary
Abstract
Introduction
Protocol
Risultati
Discussion
Divulgazioni
Acknowledgements
Materials
Riferimenti
Traduzione automatica
Please note that all translations are automatically generated. Click here for the English version.
Medicina

舌诊在传统医学中的客观化、数据分析与研究应用

Published: April 14, 2023
doi:
10.3791/65140
, , , , ,
1School of Intelligent Medicine,Chengdu University of Traditional Chinese Medicine

Summary

本研究采用U-Net等深度学习算法对舌头图像进行分割，并对比分割结果，以研究舌头诊断的客观化。

Abstract

舌诊是中医诊断的一项基本技术，通过图像处理技术客观化舌头图像的需求正在增长。本研究概述了过去十年在舌头客观化方面取得的进展，并比较了分割模型。构建了各种深度学习模型，以使用真实的舌头图像集来验证和比较算法。分析了每种模型的优缺点。结果表明，U-Net算法在精度精度（PA）、召回率和平均交集（MIoU）指标方面优于其他模型。然而，尽管在舌头图像采集和处理方面取得了重大进展，但尚未建立客观化舌头诊断的统一标准。为了促进使用移动设备捕获的舌头图像在舌头诊断对象化中的广泛应用，进一步的研究可以解决在复杂环境中捕获的舌头图像带来的挑战。

Introduction

舌观察是中国传统民族医学（TCM）中广泛使用的技术。舌头的颜色和形状可以反映身体状况和各种疾病特性、严重程度和预后。例如，在传统苗族医学中，舌头的颜色用于识别体温，例如，红色或紫色舌头表示与热有关的病理因素。在藏医中，通过观察患者的舌头，注意粘液的颜色，形状和水分来判断病情。例如，和易病患者的舌头变得红而粗糙或又黑又干1;Xieri 病2 患者出现舌黄和干燥;同时，巴达坎病3 患者舌头发白、湿润、柔软4。这些观察揭示了舌头特征与生理和病理学之间的密切关系。总体而言，舌头的状态在诊断、疾病识别和治疗效果评估中起着至关重要的作用。

同时，由于不同种族群体的生活条件和饮食习惯不同，舌头图像的差异很明显。实验室模型是在颜色测定国际标准的基础上建立的，由国际克莱里奇委员会（CIE）于1931年制定。1976年，一种颜色图案被修改并命名。Lab 颜色模型由三个元素组成：L 对应于亮度，而 a 和 b 是两个颜色通道。a 包括从深绿色（低亮度值）到灰色（中等亮度值）到亮粉色（高亮度值）的颜色;b 从亮蓝色（低亮度值）到灰色（中等亮度值）再到黄色（高亮度值）。通过比较5个民族舌色的L x a x b值，Yang等5发现苗族、回族、壮族、汉族和蒙古族的舌头图像特征彼此明显不同。例如，蒙古人的舌头颜色深，舌苔是黄色的，而苗族人的舌头是浅色的，舌苔是白色的，这表明舌头特征可以作为评估人口健康状况的诊断指标。此外，舌头图像可以作为民族医学临床研究中循证医学的评价指标。他等6以舌象作为中医诊断的基础，系统评价了周灵丹丸（CLD颗粒，用于治疗炎症和发热性疾病，包括中医季节性流感）联合中西医的安全性和有效性。研究结果确立了舌头图像作为临床研究评价指标的科学有效性。然而，传统医学从业者一般依靠主观性来观察舌头特征，评估患者的生理和病理状况，需要更精确的指标。

互联网和人工智能技术的出现为舌头诊断的数字化和客观化铺平了道路。该过程涉及使用数学模型来提供舌头图像7的定性和客观描述，反映舌头图像的内容。该过程包括几个步骤：图像采集、光学补偿、色彩校正和几何变换。然后将预处理后的图像馈送到算法模型中，用于图像定位和分割、特征提取、模式识别等。该过程的输出是对舌头图像数据的高效、精确诊断，从而达到舌诊客观化、量化化、信息化的目标8。从而达到舌诊数据高效、高精度处理的目的。本研究基于舌诊知识和深度学习技术，利用计算机算法自动将舌体和舌苔与舌头图像分离，以期为医生提取舌头定量特征，提高诊断的可靠性和一致性，为后续舌诊客观化研究提供途径9.

Protocol

本研究已获国家自然科学基金项目“基于关联分析的中医面部图像动态变化规则构建”批准。伦理批准文号为2021KL-027，伦理委员会已批准按照批准的文件开展临床研究，这些文件包括临床研究方案（2021.04.12，V2.0）、知情同意书（2021.04.12，V2.0）、受试者招募材料（2021.04.12、V2.0）、研究案例和/或病例报告、受试者日记卡和其他问卷（2021.04.12，V2.0）、临床试验参与者名单， 研究项目审批等获得?…

Representative Results

有关比较结果，请参见图 12、 图 13 和 表 1，其中本研究构建的环境使用相同的样本来训练和测试算法模型。MIoU指标：U-Net > Seg-Net > PSPNet > DeeplabV3;MPA指标：U-Net>Seg-Net>PSPNet>DeeplabV3;精准指标：U-Net > Seg-Net > DeeplabV3 > PSPNet;回想一下：U-Net > Seg-Net > PSPNet > DeeplabV3。指标值越大，分割精度越高，模型性能越好。根据指标结果可以分析出U-Net算法在…

Discussion

根据上面给出的比较结果，很明显，所考虑的四种算法的特点是不同的，它们的独特优点和缺点如下所述。U-Net结构基于全卷积网络的修改和扩展，可以通过收缩路径和对称扩展路径获得上下文信息和精确定位。通过对每个像素点进行分类，该算法实现了更高的分割精度，并更快地使用训练好的模型对图像进行分割。另一方面，Seg-Net算法由编码器和解码器的对称结构组成，具有快速适应新问题并?…

Divulgazioni

The authors have nothing to disclose.

Acknowledgements

这项工作得到了国家自然基金（批准号：82004504）、国家科技部重点研发计划（批准号：2018YFC1707606）、四川省中医药管理局（批准号：2021MS199）和国家自然基金（批准号：82174236）的支持。

Materials

CPU Intel(R) Core(TM) i7-9700K
GPU  NVIDIA GeForce RTX 3070 Ti (8192MB)
Operating systems Microsoft Windows 10 Professional Edition (64-bit)
Programming language Python
RAM 16G
DOWNLOAD MATERIALS LIST

Riferimenti

  1. Jiu, G., et al. Effects of herbal therapy on intestinal microbiota and serum metabolomics in different rat models of Mongolian medicine. Evidence-Based Complementary and Alternative. 2022, 7255780 (2022).
  2. Xi, J., Xin, Y., Teregle, Study on the correlation between the animal model of Mongolian medicine 34;Xieri disease" and serum ALT and AST. Electronic Journal of Cardiovascular Diseases in Combination of Traditional Chinese. 4 (33), 134-135 (2016).
  3. Yin, L., et al. Study on the mechanism of serum differential protein changes in bronchial asthma based on proteomics. Chinese Journal of Traditional Chinese Medicine. 47 (22), 6227-6234 (2022).
  4. Wang, X. H., Bao, L. Band Seed. The origin and development of tongue diagnosis in Mongolian medicine. Chinese Ethnic Folk Medicine. (1), 64-65 (2008).
  5. Yang, S., et al. A comparative study on the feature parameters of tongue diagnosis images in five nationalities. Chinese Journal of Traditional Chinese Medicine. 36 (11), 6428-6430 (2021).
  6. He, J. Y., et al. Efficacy and safety of Chou-Ling-Dan granules in the treatment of seasonal influenza via combining Western and traditional Chinese medicine, protocol for a multicentre, randomised controlled clinical trial. BMJ Open. 9 (4), e024800 (2019).
  7. Wang, D. J., et al. Scientific knowledge mapping and visualization analysis in the field of Chinese medicine tongue feature objectification research. World Science and Technology – Modernization of Chinese Medicine. 23 (9), 3032-3040 (2021).
  8. Yuan, S. M., Qian, P., Li, F. F. Research progress of color correction methods for tongue and face diagnosis in traditional Chinese Medicine. Chinese Journal of Traditional Chinese Medicine. 34 (9), 4183-4185 (2019).
  9. Kanawong, R., et al. Tongue image analysis and its mobile app development for health diagnosis. Advances in Experimental Medicine and Biology. 1005, 99-121 (2017).
  10. Yu, Y., et al. Semantic segmentation evaluation index and evaluation method. Computer Engineering and Application. , (2023).
  11. Sehyung, L., Negishi, M., Urakubo, H., Kasai, H., Ishii, S. Mu-net: Multi-scale U-net for two-photon microscopy image denoising and restoration. Neural Networks. 125, 92-103 (2020).
  12. Huang, X. M., et al. A review on the application of U-Net and its variants in medical image segmentation. Chinese Journal of Biomedical Engineering. 41 (5), 567-576 (2022).
  13. Lu, J. H., Xu, Y. F., Wang, Y. Q., Hao, Y. M. Research overview of tongue objectification in traditional Chinese medicine based on computer image technology. World Science and Technology – Modernization of Traditional Chinese Medicine. 24 (11), 4568-4573 (2022).
  14. Mansour, R. F., Althobaiti, M. M., Ashour, A. A. Internet of things and synergic deep learning based biomedical tongue color image analysis for disease diagnosis and classification. IEEE Access. 9, 94769-94779 (2021).
  15. Cai, Y. H., Hu, S. B., Guan, J., Zhang, X. F. Analysis of the development and application of tongue diagnosis objectification techniques in Chinese medicine. World Science and Technology – Modernization of Chinese Medicine. 23 (7), 2447-2453 (2021).
  16. Ghosh, S., Das, N., Nasipuri, M. Reshaping inputs for convolutional neural network: some common and uncommon methods. Pattern Recognition. 93, 79-94 (2019).
  17. Shang, Z. M., et al. Research progress of digital acquisition and characterization of tongue diagnosis information. Chinese Journal of Traditional Chinese Medicine. 36 (10), 6010-6013 (2021).
  18. Ning, J., Zhang, D., Wu, C., Yue, F. Automatic tongue image segmentation based on gradient vector flow and region merging. Neural Computing and Applications. 21, 1819-1826 (2012).
  19. Zheng, F., Huang, X. Y., Wang, B. L., Wang, Y. H. A method for tongue detection based on image segmentation. Journal of Xiamen University. 55 (6), 895-900 (2016).
  20. Li, Y. T., Luo, Y. S., Zhu, Z. M. Deep learning-based tongue feature analysis. Computer Science. 47 (11), 148-158 (2020).
  21. Wang, Y. D., Sun, C. H., Cui, J. L., Wu, X. R., Qin, Y. X. Research on deep learning-based tongue fissure segmentation algorithm. World Science and Technology – Modernization of Chinese Medicine. 23 (9), 3065-3073 (2021).
  22. Wang, X. M., Wang, R. Y., Guo, D., Lu, S. Z., Zhou, P. Research on the identification method of tongue punctures based on auxiliary light source. Journal of Sensing Technology. 29 (10), 1553-1559 (2016).
  23. Yang, J. D., Zhang, P. A fully connected neural network based on migration learning for tongue image classification. Journal of the Second Military Medical University. 39 (8), 897-902 (2018).
  24. Song, C., Wang, B., Xu, J. T. Research on tongue feature classification method based on deep migration learning. Computer Engineering and Science. 43 (8), 1488-1496 (2021).
  25. Ding, H. J., He, J. C. Study on modern techniques and methods of tongue diagnosis. Shi Zhen Chinese Medicine. 21 (5), 1230-1232 (2010).
  26. Zhai, P. B., et al. A multi-stage tongue image classification algorithm incorporating attention mechanism. Computer Engineering and Design. 42 (6), 1606-1613 (2021).
  27. Hou, Y. S. A new clustering analysis algorithm based on deep learning. Journal of Xinxiang University. 35 (12), 4 (2018).
  28. Shi, Y. L., et al. A decision tree algorithm for classification of non-small cell lung cancer evidence based on tongue and pulse data. World Science and Technology – Modernization of Chinese Medicine. 24 (7), 2766-2775 (2022).
  29. Li, J., Hu, X. J., Zhou, C. L., Xu, J. T. Study on the feature analysis and diagnosis model of diabetic tongue based on random forest algorithm. Chinese Journal of Traditional Chinese Medicine. 37 (3), 1639-1643 (2022).
  30. Chen, C. H., et al. The characteristics of the combination of the four diagnostic methods of traditional Chinese medicine from the perspective of the differences between Chinese and Western medical diagnosis methods. Journal of Guangzhou University of Traditional Chinese Medicine. 28 (3), 332-334 (2011).
  31. Chen, R., et al. Correlation analysis of tongue image and western medicine typing in 382 patients with novel coronavirus pneumonia based on Lab colour model and imaging histology. Chinese Journal of Traditional Chinese Medicine. 36 (12), 7010-7014 (2021).
  32. Ju, J. W., Jung, H., Lee, Y. J., Mun, S. W., Lee, J. H. Semantic segmentation dataset for AI-based quantification of clean mucosa in capsule endoscopy. Medicina. 58 (3), 397 (2022).
  33. Wu, X., et al. A review of research on deep learning in tongue image classification. Computer Science and Exploration. , 1-23 (2022).

Tags

Tongue Image SegmentationDeep LearningTraditional MedicineData AnalysisLingual Face Diagnostic InstrumentImage AcquisitionManual ScreeningLabelMeImage Segmentation Algorithm
check_url/it/65140?article_type=t

Play Video
PDF
DOI
DOWNLOAD MATERIALS LIST
Citazione di questo articolo
Feng, L., Xiao, W., Wen, C., Deng, Q., Guo, J., Song, H. Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application. J. Vis. Exp. (194), e65140, doi:10.3791/65140 (2023).
Scarica citazione
Ristampe e autorizzazioni

View Video

✖

To prove you're not a robot, please enter the text in the image below

CAPTCHA Image
Play CAPTCHA Audio
Refresh Image