Naimin Li

Computerized tongue diagnosis based on Bayesian networks

Tongue diagnosis is an important diagnostic method in traditional Chinese medicine (TCM). However, due to its qualitative, subjective and experience-based nature, traditional tongue diagnosis has a very limited application in clinical medicine. Moreover, traditional tongue diagnosis is always concerned with the identification of syndromes rather than with the connection between tongue abnormal appearances and diseases. This is not well understood in Western medicine, thus greatly obstruct its wider use in the world. In this paper, we present a novel computerized tongue inspection method aiming to address these problems. First, two kinds of quantitative features, chromatic and textural measures, are extracted from tongue images by using popular digital image processing techniques. Then, Bayesian networks are employed to model the relationship between these quantitative features and diseases. The effectiveness of the method is tested on a group of 455 patients affected by 13 common diseases as well as other 70 healthy volunteers, and the diagnostic results predicted by the previously trained Bayesian network classifiers are reported.

A Novel Breath Analysis System Based on Electronic Olfaction

Certain gases in the breath are known to be indicators of the presence of diseases and clinical conditions. These gases have been identified as biomarkers using equipments, such as gas chromatography and electronic nose (e-nose). GC is very accurate but is expensive, time consuming, and nonportable. E-nose has the advantages of low cost and easy operation, but is not particular for analyzing breath odor, and hence, has a limited application in diseases diagnosis. This paper proposes a novel system that is special for breath analysis. We selected chemical sensors that are sensitive to the biomarkers and compositions in human breath, developed the system, and introduced the odor signal preprocessing and classification method. To evaluate the system performance, we captured breath samples from healthy persons and patients known to be afflicted with diabetes, renal disease, and airway inflammation, respectively, and conducted experiments on medical treatment evaluation and disease identification. The results show that the system is not only able to distinguish between breath samples from subjects suffering from various diseases or conditions (diabetes, renal disease, and airway inflammation) and breath samples from healthy subjects, but in the case of renal failure is also helpful in evaluating the efficacy of hemodialysis (treatment for renal failure).

Tongue shape classification by geometric features

Traditional Chinese Medicine diagnoses a wide range of health conditions by examining features of the tongue, including its shape. This paper presents a classification approach for automatically recognizing and analyzing tongue shapes based on geometric features. The approach corrects the tongue deflection by applying three geometric criteria and then classifies tongue shapes according to seven geometric features defined using various measurements of length, area and angle of the tongue. To establish a measurable and machine readable relationship between expert human judgments and the machine classifications of tongue shapes, we use a decision support tool, Analytic Hierarchy Process (AHP), to weight the relative influences of the various length/area/angle factors used in classifying a tongue, and then apply a fuzzy fusion framework that combines seven AHP modules, one for each tongue shape, to represent the uncertainty and imprecision between these quantitative features and tongue shape classes. Experimental results show that the proposed shape correction method reduces the deflection of tongue shapes and that our shape classification approach, tested on a total of 362 tongue samples, achieved an accuracy of 90.3%, making it more accurate than either KNN or LDA.

Pulse images recognition using fuzzy neural network

The automatic recognition of pulse images is the key in the research of computerized pulse diagnosis. In order to automatically differentiate the pulse patterns by using small samples in pulse diagnosis, a fuzzy neural network for classifying pulse images based on the knowledge of experts in traditional Chinese pulse diagnosis was designed. The designed classifier can make hard decision and soft decision for identifying 16 patterns of pulse images at the accuracy of 90.25%, which is better than the results that are achieved by back propagation neural network.

Computerized Diagnosis from Tongue Appearance Using Quantitative Feature Classification

This study investigates relationships between diseases and the appearance of the human tongue in terms of quantitative features. The experimental samples are digital tongue images captured from three groups of candidates: one group in normal health, one suffering with appendicitis, and a third suffering with pancreatitis. For the purposes of diagnostic classification, we first extract chromatic and textural measurements from original tongue images. A feature selection procedure then identifies the measures most relevant to the classifications, based on which of the three tongue image categories are clearly separated. This study validates the use of tongue inspection by means of quantitative feature classification in medical diagnosis.

A wavelet packet based pulse waveform analysis for cholecystitis and nephrotic syndrome diagnosis

Traditional Chinese Pulse Diagnosis (TCPD), one of the four diagnostic methods of Traditional Chinese Medicine (TCM), had been proved to be clinically valid in Chinese Medicine history. Different from most previous work which focused on the diagnosis of cardiovascular diseases, this paper further investigated the possibility of diagnosing cholecystitis and nephrotic syndrome using the pulse waveform data. After the pre-processing, the pulse waveform signals were decomposed into a given level by Wavelet Packet Transform and the best basis was picked out by Shannon entropy criterion. Then, subband energies contained in the best basis were extracted as features and the support vector machine classifiers were trained. Experimental results indicated that the proposed method can effectively discriminate these two kinds of diseases.

Combination of Heterogeneous Features for Wrist Pulse Blood Flow Signal Diagnosis via Multiple Kernel Learning

Wrist pulse signal is of great importance in the analysis of the health status and pathologic changes of a person. A number of feature extraction methods have been proposed to extract linear and nonlinear, and time and frequency features of wrist pulse signal. These features are heterogeneous in nature and are likely to contain complementary information, which highlights the need for the integration of heterogeneous features for pulse classification and diagnosis. In this paper, we propose a novel effective method to classify the wrist pulse blood flow signals by using the multiple kernel learning (MKL) algorithm to combine multiple types of features. In the proposed method, seven types of features are first extracted from the wrist pulse blood flow signals using the state-of-the-art pulse feature extraction methods, and are then fed to an efficient MKL method, SimpleMKL, to combine heterogeneous features for more effective classification. Experimental results show that the proposed method is promising in integrating multiple types of pulse features to further enhance the classification performance.

Pulse Contour Variability Before and After Exercise

This paper compares the radial artery pulses of 105 young graduate students. The radial artery pulses after performing progressive ergometer for five minutes are different from those at rest. All the pulses become floating and fast. The contours of pulses have three kinds of variability. The incisures of 39 subjects become especially low; sometimes the incisures are lower than the onset of pulse waveform. The tidal waves and dicrotic waves of 32 subjects become higher. The pulses of 34 subjects become smooth. Their incisures and dicrotic waves become lower. These changes can instruct the exercise and training of the young students and athletes

Classification of pulse waveforms using edit distance with real penalty

Advances in sensor and signal processing techniques have provided effective tools for quantitative research in traditional Chinese pulse diagnosis (TCPD). Because of the inevitable intraclass variation of pulse patterns, the automatic classification of pulse waveforms has remained a difficult problem. In this paper, by referring to the edit distance with real penalty (ERP) and the recent progress in -nearest neighbors (KNN) classifiers, we propose two novel ERP-based KNN classifiers. Taking advantage of the metric property of ERP, we first develop an ERP-induced inner product and a Gaussian ERP kernel, then embed them into difference-weighted KNN classifiers, and finally develop two novel classifiers for pulse waveform classification. The experimental results show that the proposed classifiers are effective for accurate classification of pulse waveform.

Diabetes identification and classification by means of a breath analysis system

This article proposes a breath analysis system that makes use of chemical sensors to detect acetone in human breath, and hence detect the diabetes and measure the blood glucose levels of diabetics. We captured the breath samples from healthy persons and patients known to be afflicted with diabetes and conducted experiments on disease identification and simultaneous blood glucose measurement. SVM classifier was used to identify diabetes from healthy samples and three models were built to fit the curves that can represent the blood glucose levels. The results show that the system is not only able to distinguish between breath samples from patients with diabetes and healthy subjects, but also to represent the fluctuation of blood sugar of diabetics and therefore to be an evaluation tool for monitoring the blood glucose of diabetes.