Liping Tu

A medical image color correction method base on supervised color constancy

Objective: This paper aims to establish an acquisition and analysis method for medical image in the indoor natural light conditions; Method: According to the color characteristics of the medical image and the constancy of supervised color, we make the color codes which are used for a* b* value correction, and gray codes which are used for L* value correction. Then, we carry on topological subdivision to the one-dimensional L* space and the two-dimensional a*b* space respectively. We establish the mapping function ldquodecomposition-projection-restorationrdquo, and ran the linear mapping to the one-dimensional L* space and triangle mapping to the two-dimensional a*b* space. We correct the 198 color blocks in 22 pictures. Result: Compared with the standard value, DeltaL*, DeltaC*, DeltaE decreased significantly (P<0.01), imagepsilas color difference is reduced, color saturation and the value is closer to standard value. Conclusion: TRM(topology resolve-map model) method can significantly adjust the color error of tongue images under the indoor natural light condition and easy to use, and has a good performance of color correction. It is feasible to apply this method to the color correction of the medical images.

A Diagnostic Method Based on Tongue Imaging Morphology

The tongue morphology analysis is an important part of imaging diagnosis of tongue inspection. This paper studies and analyzes the tongue shape to establish a kind of tongue diagnosis method based on tongue images. CCD devices were employed to acquire the front and lateral images of tongue, and then measure the tongues length (L), width (K) and height (H). Furthermore, in 500 testees with possible disorders, the coefficient of variation(CV) was corrected and adjusted to establish an optimum formula between the body surface area (Mt) and the sum of the width of the tongue (K) and height (H). Clinical studies show that this tongue judgment method is of high accuracy. The accuracy rate for fat tongue and thin tongue reaches 93.40% and 88.57% respectively. In addition, this method has been proven to be valuable in the diagnosis of diabetes mellitus, hypertension, chronic gastritis and hyperthyroidism.

Diagnostic Method of Diabetes Based on Support Vector Machine and Tongue Images

Objective. The purpose of this research is to develop a diagnostic method of diabetes based on standardized tongue image using support vector machine (SVM). Methods. Tongue images of 296 diabetic subjects and 531 nondiabetic subjects were collected by the TDA-1 digital tongue instrument. Tongue body and tongue coating were separated by the division-merging method and chrominance-threshold method. With extracted color and texture features of the tongue image as input variables, the diagnostic model of diabetes with SVM was trained. After optimizing the combination of SVM kernel parameters and input variables, the influences of the combinations on the model were analyzed. Results. After normalizing parameters of tongue images, the accuracy rate of diabetes predication was increased from 77.83% to 78.77%. The accuracy rate and area under curve (AUC) were not reduced after reducing the dimensions of tongue features with principal component analysis (PCA), while substantially saving the training time. During the training for selecting SVM parameters by genetic algorithm (GA), the accuracy rate of cross-validation was grown from 72% or so to 83.06%. Finally, we compare with several state-of-the-art algorithms, and experimental results show that our algorithm has the best predictive accuracy. Conclusions. The diagnostic method of diabetes on the basis of tongue images in Traditional Chinese Medicine (TCM) is of great value, indicating the feasibility of digitalized tongue diagnosis.

Tongue Images Classification Based on Constrained High Dispersal Network

Computer aided tongue diagnosis has a great potential to play important roles in traditional Chinese medicine (TCM). However, the majority of the existing tongue image analyses and classification methods are based on the low-level features, which may not provide a holistic view of the tongue. Inspired by deep convolutional neural network (CNN), we propose a novel feature extraction framework called constrained high dispersal neural networks (CHDNet) to extract unbiased features and reduce human labor for tongue diagnosis in TCM. Previous CNN models have mostly focused on learning convolutional filters and adapting weights between them, but these models have two major issues: redundancy and insufficient capability in handling unbalanced sample distribution. We introduce high dispersal and local response normalization operation to address the issue of redundancy. We also add multiscale feature analysis to avoid the problem of sensitivity to deformation. Our proposed CHDNet learns high-level features and provides more classification information during training time, which may result in higher accuracy when predicting testing samples. We tested the proposed method on a set of 267 gastritis patients and a control group of 48 healthy volunteers. Test results show that CHDNet is a promising method in tongue image classification for the TCM study.

The Classification of Tongue Colors with Standardized Acquisition and ICC Profile Correction in Traditional Chinese Medicine

Background and Goal. The application of digital image processing techniques and machine learning methods in tongue image classification in Traditional Chinese Medicine (TCM) has been widely studied nowadays. However, it is difficult for the outcomes to generalize because of lack of color reproducibility and image standardization. Our study aims at the exploration of tongue colors classification with a standardized tongue image acquisition process and color correction. Methods. Three traditional Chinese medical experts are chosen to identify the selected tongue pictures taken by the TDA-1 tongue imaging device in TIFF format through ICC profile correction. Then we compare the mean value of L * a * b * of different tongue colors and evaluate the effect of the tongue color classification by machine learning methods. Results. The L * a * b * values of the five tongue colors are statistically different. Random forest method has a better performance than SVM in classification. SMOTE algorithm can increase classification accuracy by solving the imbalance of the varied color samples. Conclusions. At the premise of standardized tongue acquisition and color reproduction, preliminary objectification of tongue color classification in Traditional Chinese Medicine (TCM) is feasible.

Pulse Wave Cycle Features Analysis of Different Blood Pressure Grades in the Elderly

Background and Objective The same range of blood pressure values may reflect different vascular functions, especially in the elderly. Therefore, a single blood pressure value may not comprehensively reveal cardiovascular function. This study focused on identifying pulse wave features in the elderly that can be used to show functional differences when blood pressure values are in the same range. Methods First, pulse data were preprocessed and pulse cycles were segmented. Second, time domain, higher-order statistics, and energy features of wavelet packet decomposition coefficients were extracted. Finally, useful pulse wave features were evaluated using a feature selection and classifier design. Results A total of 6,075 pulse wave cycles were grouped into 3 types according to different blood pressure levels and each group was divided into 2 categories according to a history of hypertension. The classification accuracy of feature selection in the 3 groups was 97.91%, 95.24%, and 92.28%, respectively. Conclusion Selected features could be appropriately used to analyze cardiovascular function in the elderly and can serve as the basis for research on a cardiovascular risk assessment model based on Traditional Chinese Medicine pulse diagnosis.

Analysis of Pulse Signals Based on Array Pulse Volume

ObjectiveTo collect and analyze multi-dimensional pulse diagram features with the array sensor of a pressure profile system (PPS) and study the characteristic parameters of the new multi-dimensional pulse diagram by pulse diagram analysis technology.MethodsThe pulse signals at the Guan position of left wrist were acquired from 105 volunteers at the Shanghai University of Traditional Chinese Medicine. We obtained the pulse data using an array sensor with 3×4 channels. Three dimensional pulse diagrams were constructed for the validated pulse data, and the array pulse volume (APV) parameter was computed by a linear interpolation algorithm. The APV differences among normal pulse (NP), wiry pulse (WP) and slippery pulse (SP) were analyzed using one-way analysis of variance. The coefficients of variation (CV) were calculated for WP, SP and NP.ResultsThe APV difference between WP and NP in the 105 volunteers was statistically significant (6.26±0.28 vs. 6.04±0.36, P=0.048), as well as the difference between WP and SP (6.26±0.28 vs. 6.07±0.46, P=0.049). However, no statistically significant difference was found between NP and SP (P=0.75). WP showed a similar CV (4.47%) to those of NP (5.96%) and SP (7.58%).ConclusionThe new parameter APV could differentiate between NP or SP and WP. Accordingly, APV could be considered an useful parameter for the analysis of array pulse diagrams in Chinese medicine.

A deep tongue image features analysis model for medical application

With the improvement of peoples living standards, there is no doubt that people are paying more and more attention to their health. However, shortage of medical resources is a critical global problem. As a result, an intelligent prognostics system has a great potential to play important roles in computer aided diagnosis. Numerous papers reported that tongue features have been closely related to a humans state. Among them, the majority of the existing tongue image analyses and classification methods are based on the low-level features, which may not provide a holistic view of the tongue. Inspired by a deep convolutional neural network (CNN), we propose a deep tongue image feature analysis system to extract unbiased features and reduce human labor for tongue diagnosis. With the unbalanced sample distribution, it is hard to form a balanced classification model based on feature representations obtained by existing low-level and high-level methods. Our proposed deep tongue image feature analysis model learns high-level features and provide more classification information during training time, which may result in higher accuracy when predicting testing samples. We tested the proposed system on a set of 267 gastritis patients, and a control group of 48 healthy volunteers (labeled according to Western medical practices). Test results show that the proposed deep tongue image feature analysis model can classify a given tongue image into healthy and diseased state with an average accuracy of 91.49%, which demonstrates the relationship between human bodys state and its deep tongue image features.

A Shiatsu pulse sensor calibration method and application to three-part pulse wave collection

Radial artery pulse wave is a very important biological signal in the Traditional Chinese Medical (TCM) theory. It also plays an important role in the field of health informatics. In this paper, we proposed a new shiatsu three-part pulse collection method based on TCM to collect three-part pulses objectively. We established a feasible method for three-part pulse collection, and provided an objective process for radial artery pulse collection. The result of this research can be applied to disease diagnosis and health informatics.

Recognize and analysis the healthy person's facial color diagnosis

We fetch facial digital vision information from 98 healthy persons using digital camera. With functional help of NIKON CAPTURE NXs software point U key technology, we catch black, white and neutral color control points. We correct the color differences of facial digital vision with self-improved facial functional software. According to the principle of Adaboost objective category, we improve the software. Under Traditional Chinese Medicine (TCM), facial information was cut into 5 divisions which are forehead, nose, left malar, right malar and chin. We do the CIE1976L*a*b* color quantity analysis using after we take 8 color pieces with size 5*5 from each part. The experiment result shows that there are differences of value of L*, a* and b* from different part of healthy persons face. That means there are color differences from different part of face. With the figures of L*, a* and b* from different face part, it generally follows the TCM clinic fact. This task technologically found the patients facial inspection in future.