Deep Learning Multi-label Tongue Image Analysis and Its Application in a Population Undergoing Routine Medical Checkup.

Background: Research on intelligent tongue diagnosis is a main direction in the modernization of tongue diagnosis technology. Identification of tongue shape and texture features is a difficult task for tongue diagnosis in traditional Chinese medicine (TCM). This study aimed to explore the application of deep learning techniques in tongue image analyses. Methods: A total of 8676 tongue images were annotated by clinical experts, into seven categories, including the fissured tongue, tooth-marked tongue, stasis tongue, spotted tongue, greasy coating, peeled coating, and rotten coating. Based on the labeled tongue images, the deep learning model faster region-based convolutional neural networks (Faster R-CNN) was utilized to classify tongue images. Four performance indices, i.e., accuracy, recall, precision, and F1-score, were selected to evaluate the model. Also, we applied it to analyze tongue image features of 3601 medical checkup participants in order to explore gender and age factors and the correlations among tongue features in diseases through complex networks. Results: The average accuracy, recall, precision, and F1-score of our model achieved 90.67%, 91.25%, 99.28%, and 95.00%, respectively. Over the tongue images from the medical checkup population, the model Faster R-CNN detected 41.49% fissured tongue images, 37.16% tooth-marked tongue images, 29.66% greasy coating images, 18.66% spotted tongue images, 9.97% stasis tongue images, 3.97% peeled coating images, and 1.22% rotten coating images. There were significant differences in the incidence of the fissured tongue, tooth-marked tongue, spotted tongue, and greasy coating among age and gender. Complex networks revealed that fissured tongue and tooth-marked were closely related to hypertension, dyslipidemia, overweight and nonalcoholic fatty liver disease (NAFLD), and a greasy coating tongue was associated with hypertension and overweight. Conclusion: The model Faster R-CNN shows good performance in the tongue image classification. And we have preliminarily revealed the relationship between tongue features and gender, age, and metabolic diseases in a medical checkup population.

A multi-step approach for tongue image classification in patients with diabetes.

BACKGROUND: In China, diabetes is a common, high-incidence chronic disease. Diabetes has become a severe public health problem. However, the current diagnosis and treatment methods are difficult to control the progress of diabetes. Traditional Chinese Medicine (TCM) has become an option for the treatment of diabetes due to its low cost, good curative effect, and good accessibility. OBJECTIVE: Based on the tongue images data to realize the fine classification of the diabetic population, provide a diagnostic basis for the formulation of individualized treatment plans for diabetes, ensure the accuracy and consistency of the TCM diagnosis, and promote the objective and standardized development of TCM diagnosis. METHODS: We use the TFDA-1 tongue examination instrument to collect the tongue images of the subjects. Tongue Diagnosis Analysis System (TDAS) is used to extract the TDAS features of the tongue images. Vector Quantized Variational Autoencoder (VQ-VAE) extracts VQ-VAE features from tongue images. Based on VQ-VAE features, K-means clustering tongue images. TDAS features are used to describe the differences between clusters. Vision Transformer (ViT) combined with Grad-weighted Class Activation Mapping (Grad-CAM) is used to verify the clustering results and calculate positioning diagnostic information. RESULTS: Based on VQ-VAE features, K-means divides the diabetic population into 4 clusters with clear boundaries. The silhouette, calinski harabasz, and davies bouldin scores are 0.391, 673.256, and 0.809, respectively. Cluster 1 had the highest Tongue Body L (TB-L) and Tongue Coating L (TC-L) and the lowest Tongue Coating Angular second moment (TC-ASM), with a pale red tongue and white coating. Cluster 2 had the highest TC-b with a yellow tongue coating. Cluster 3 had the highest TB-a with a red tongue. Group 4 had the lowest TB-L, TC-L, and TB-b and the highest Per-all with a purple tongue and the largest tongue coating area. ViT verifies the clustering results of K-means, the highest Top-1 Classification Accuracy (CA) is 87.8%, and the average CA is 84.4%. CONCLUSIONS: The study organically combined unsupervised learning, self-supervised learning, and supervised learning and designed a complete diabetic tongue image classification method. This method does not rely on human intervention, makes decisions based entirely on tongue image data, and achieves state-of-the-art results. Our research will help TCM deeply participate in the individualized treatment of diabetes and provide new ideas for promoting the standardization of TCM diagnosis.

Research of the Distribution of Tongue Features of Diabetic Population Based on Unsupervised Learning Technology.

Background: The prevalence of diabetes increases year by year, posing a severe threat to human health. Current treatments are difficult to prevent the progression of diabetes and its complications. It is imperative to carry out individualized treatment of diabetes, but current diagnostic methods are difficult to specify an individualized treatment plan. Objective: Clarify the distribution law of tongue features of the diabetic population, and provide the diagnostic basis for individualized treatment of traditional Chinese medicine (TCM) in the treatment of diabetes. Methods: We use the TFDA-1 tongue diagnosis instrument to collect tongue images of people with diabetes and accurately calculate the color features, texture features, and tongue coating ratio features through the Tongue Diagnosis Analysis System (TDAS). Then, we used K-means and Self-organizing Maps (SOM) networks to analyze the distribution of tongue features in diabetic people. Statistical analysis of TDAS features was used to identify differences between clusters. Results: The silhouette coefficient of the K-means clustering result is 0.194, and the silhouette coefficient of the SOM clustering result is 0.127. SOM Cluster 3 and Cluster 4 are derived from K-means Cluster 1, and the intersections account for (76.7% 97.5%) and (22.3% and 70.4%), respectively. K-means Cluster 2 and SOM Cluster 1 are highly overlapping, and the intersection accounts for the ratios of 66.9% and 95.0%. K-means Cluster 3 and SOM Cluster 2 are highly overlaid, and the intersection ratio is 94.1% and 82.1%. For the clustering results of K-means, TB-a and TC-a of Cluster 3 are the highest (P < 0.001), TB-a of Cluster 2 is the lowest (P < 0.001), and TB-a of Cluster 1 is between Cluster 2 and Cluster 3 (P < 0.001). Cluster 1 has the highest TB-b and TC-b (P < 0.001), Cluster 2 has the lowest TB-b and TC-b (P < 0.001), and TB-b and TC-b of Cluster 3 are between Cluster 1 and Cluster 2 (P < 0.001). Cluster 1 has the highest TB-ASM and TC-ASM (P < 0.001), Cluster 3 has the lowest TB-ASM and TC-ASM (P < 0.001), and TB-ASM and TC-ASM of Cluster 2 are between the Cluster 1 and Cluster 3 (P < 0.001). CON, ENT, and MEAN show the opposite trend. Cluster 2 had the highest Per-all (P < 0.001). SOM divides K-means Cluster 1 into two categories. There is almost no difference in texture features between Cluster 3 and Cluster 4 in the SOM clustering results. Cluster 3's TB-L, TC-L, and Per-all are lower than Cluster 4 (P < 0.001), Cluster 3's TB-a, TC-a, TB-b, TC-b, and Per-part are higher than Cluster 4 (P < 0.001). Conclusions: The precise tongue image features calculated by TDAS are the basis for characterizing the disease state of diabetic people. Unsupervised learning technology combined with statistical analysis is an important means to discover subtle changes in the tongue features of diabetic people. The machine vision analysis method based on unsupervised machine learning technology realizes the classification of the diabetic population based on fine tongue features. It provides a diagnostic basis for the designated diabetes TCM treatment plan.

A tongue features fusion approach to predicting prediabetes and diabetes with machine learning.

BACKGROUND: Diabetics has become a serious public health burden in China. Multiple complications appear with the progression of diabetics pose a serious threat to the quality of human life and health. We can prevent the progression of prediabetics to diabetics and delay the progression to diabetics by early identification of diabetics and prediabetics and timely intervention, which have positive significance for improving public health. OBJECTIVE: Using machine learning techniques, we establish the noninvasive diabetics risk prediction model based on tongue features fusion and predict the risk of prediabetics and diabetics. METHODS: Applying the type TFDA-1 Tongue Diagnosis Instrument, we collect tongue images, extract tongue features including color and texture features using TDAS, and extract the advanced tongue features with ResNet-50, achieve the fusion of the two features with GA_XGBT, finally establish the noninvasive diabetics risk prediction model and evaluate the performance of testing effectiveness. RESULTS: Cross-validation suggests the best performance of GA_XGBT model with fusion features, whose average CA is 0.821, the average AUROC is 0.924, the average AUPRC is 0.856, the average Precision is 0.834, the average Recall is 0.822, the average F1-score is 0.813. Test set suggests the best testing performance of GA_XGBT model, whose average CA is 0.81, the average AUROC is 0.918, the average AUPRC is 0.839, the average Precision is 0.821, the average Recall is 0.81, the average F1-score is 0.796. When we test prediabetics with GA_XGBT model, we find that the AUROC is 0.914, the Precision is 0.69, the Recall is 0.952, the F1-score is 0.8. When we test diabetics with GA_XGBT model, we find that the AUROC is 0.984, the Precision is 0.929, the Recall is 0.951, the F1-score is 0.94. CONCLUSIONS: Based on tongue features, the study uses classical machine learning algorithm and deep learning algorithm to maximum the respective advantages. We combine the prior knowledge and potential features together, establish the noninvasive diabetics risk prediction model with features fusion algorithm, and detect prediabetics and diabetics noninvasively. Our study presents a feasible method for establishing the association between diabetics and the tongue image information and prove that tongue image information is a potential marker which facilitates effective early diagnosis of prediabetics and diabetics.

Effect of pulsatile perfusion during cardiopulmonary bypass in terms of radial artery sphygmogram.

OBJECTIVE: To investigate a quantitative method for using radial artery pulse waveforms to assess the effect of pulsatile flow during cardiopulmonary bypass (CPB). METHODS: A total of 34 adults with heart disease who underwent open-heart surgery between April 2010 and January 2011 were randomized into a pulsatile perfusion group (n = 17) and a non-pulsatile perfusion group (n = 17). Radial arterial pulse waveforms of pulsatile and non-pulsatile perfusion patients were observed and compared before and during CDB. RESULTS: No pulse waveform could be detected at patients' radial artery in both groups when the aorta was cross-clamped. Pulse waveforms could be detected at pulsatile perfusion patients' radial artery, but could not be detected at non-pulsatile perfusion patients' radial artery during CPB. Additionally, patients' pulse waveforms during pulsatile perfusion were lower than those before the operation. CONCLUSION: Our findings indicate that radial artery sphygmogram can be used as a valid indicator to evaluate the effectiveness of pulsatile perfusion during CPB.

Graph theoretical analysis of EEG functional connectivity during music perception.

The present study evaluated the effect of music on large-scale structure of functional brain networks using graph theoretical concepts. While most studies on music perception used Western music as an acoustic stimulus, Guqin music, representative of Eastern music, was selected for this experiment to increase our knowledge of music perception. Electroencephalography (EEG) was recorded from non-musician volunteers in three conditions: Guqin music, noise and silence backgrounds. Phase coherence was calculated in the alpha band and between all pairs of EEG channels to construct correlation matrices. Each resulting matrix was converted into a weighted graph using a threshold, and two network measures: the clustering coefficient and characteristic path length were calculated. Music perception was found to display a higher level mean phase coherence. Over the whole range of thresholds, the clustering coefficient was larger while listening to music, whereas the path length was smaller. Networks in music background still had a shorter characteristic path length even after the correction for differences in mean synchronization level among background conditions. This topological change indicated a more optimal structure under music perception. Thus, prominent small-world properties are confirmed in functional brain networks. Furthermore, music perception shows an increase of functional connectivity and an enhancement of small-world network organizations.

[Ashi points, Ashi method and pressing reaction].

The nomination of Ashi points was reviewed, and the meaning of Ashi method was discussed in this article. On the base of further study on Huangdi Neijing (The Yellow Emperor's Canon of Internal Medicine), the general meaning of palpation at acupoints, meridians and collaterals to the process of acupoint locating were expounded. The concept of pressing reaction was proposed as well. It is held that Ashi points are a category of acupoints without specific names and definite locations. They are a kind of manifestation of reactions of acupoints, meridians and collaterals, which embody their dynamic features. Pressing reaction mainly manifested by sensations of comfort, pain, and moreover, relieving of the primary symptoms. It is the most basic evidence for us to estimate Ashi points.

Crossmodal effects of Guqin and piano music on selective attention: an event-related potential study.

To compare the effects of music from different cultural environments (Guqin: Chinese music; piano: Western music) on crossmodal selective attention, behavioral and event-related potential (ERP) data in a standard two-stimulus visual oddball task were recorded from Chinese subjects in three conditions: silence, Guqin music or piano music background. Visual task data were then compared with auditory task data collected previously. In contrast with the results of the auditory task, the early (N1) and late (P300) stages exhibited no differences between Guqin and piano backgrounds during the visual task. Taking our previous study and this study together, we can conclude that: although the cultural-familiar music influenced selective attention both in the early and late stages, these effects appeared only within a sensory modality (auditory) but not in cross-sensory modalities (visual). Thus, the musical cultural factor is more obvious in intramodal than in crossmodal selective attention.

The influence of Mozart's sonata K.448 on visual attention: an ERPs study.

In the present study, the effects of Mozart's sonata K.448 on voluntary and involuntary attention were investigated by recording and analyzing behavioral and event-related potentials (ERPs) data in a three-stimulus visual oddball task. P3a (related to involuntary attention) and P3b (related to voluntary attention) were analyzed. The "Mozart effect" was showed on ERP but not on behavioral data. This study replicated the previous results of Mozart effect on voluntary attention: the P3b latency was influenced by Mozart's sonata K.448. But no change of P3a latency was induced by this music. At the same time, decreased P3a and P3b amplitudes in music condition were found. We interpret this change as positive "Mozart effect" on involuntary attention (P3a) and negative "Mozart effect" on voluntary attention (P3b). We conclude that Mozart's sonata K.448 has shown certain effects on both involuntary attention and voluntary attention in our study, but their effects work on different mechanisms.

[Correlation analysis of infrared radiant intensity between hyperplastic breast tissue and tongue surface in patients with hyperplasia of mammary glands].

OBJECTIVE: To establish a supplementary diagnostic indicator (infrared radiant intensity) in tongue diagnosis of traditional Chinese medicine (TCM) in patients with hyperplasia of mammary glands through correlation analysis of infrared radiant intensity between hyperplastic breast tissue and tongue surface. METHODS: Infrared radiant intensity of the hyperplastic breast tissue and different points on tongue surface in 20 cases of hyperplasia of mammary glands with liver-energy stagnation and phlegm retention syndrome and 16 cases of hyperplasia of mammary glands with irregular thoroughfare and conception vessels syndrome were measured with external infrared spectrometer PHE-201 made by Shanghai Institute of Technical Physics. Correlation of infrared radiant intensity between the hyperplastic breast tissue and the different points on tongue surface was assessed by using bivariate correlation analysis. RESULTS: The results showed that the numbers of positive correlated wave bands of infrared radiant intensity between the hyperplastic breast tissue and different detected points on tongue surface in the patients with liver-energy stagnation and phlegm retention syndrome and irregular thoroughfare and conception vessels syndrome were 127 (83.55%) and 71 (46.71%), respectively. Infrared radiant intensity between the hyperplastic breast tissue and the tongue surface had a positive correlation. CONCLUSION: Infrared radiant intensity can be used as one of supplementary diagnostic indicators in TCM tongue diagnosis of hyperplasia of mammary glands.

[Current progress of research on measuring sphygmus information].

Based on summarizing the physical features to describe the sphygmus information in traditional Chinese medicine (TCM), this paper surveys the methodologies of sphygmus measuring, such as single-probe measurement and multi-probe measurement based on pressure sensors, as well as measurement methods based on non-pressure sensors. It is concluded that to achieve the comprehensiveness on the sphygmus information measurement and analysis, researchers need to do further studies of the underlying mechanism and the information properties of the sphygmus. In addition, the sphygmus system should be modeled physically and mathematically.

[Progress and prospects of research on information processing techniques for intelligent diagnosis of traditional Chinese medicine].

Information processing for intelligent diagnosis of traditional Chinese medicine (TCM), an important part of the modernization of Chinese medicine, attracts world wide attention from the science circle. This article presents a systematic introduction to the development of information technology, especially the processing of pulse and tongue images and systems of computer-aided Chinese medical diagnosis. Furthermore, it points out four essential areas of future research, including epistemic logic system of syndrome differentiation, system construction technology, data miming technology and information acquisition and analysis in TCM diagnosis.

[Evaluation of therapeutic effects of Chinese materia medica by tongue image analysis software 1.0 based on tongue colors].

OBJECTIVE: To evaluate the therapeutic effects of Chinese materia medica in treating patients with different syndromes by tongue image analysis software 1.0 based on tongue colors, and to discuss the feasibility of applying this computer science-based techniques into drug evaluation. METHODS: The tongue colors and the areas of tongue fur were examined and analyzed by the tongue image analysis software 1.0 in healthy persons and the patients with different syndromes before and after treatment. The parameters of tongue colors consisted of the followings: the hue (H), the lightness (L), the saturation (S), and the values of red (R), green (G) and blue (B). RESULTS: Obvious differences could be revealed in tongue color index between the healthy persons and the patients in five groups of different syndromes. There also existed some significant differences in those index between patients before and after treatment. CONCLUSION: The tongue image analysis software 1.0 based on tongue colors is helpful to evaluate the therapeutic effects of Chinese materia medica.