Search

As a complementary and alternative medicine in medical field, traditional Chinese medicine (TCM) has drawn great attention in the domestic field and overseas. In practice, TCM provides a quite distinct methodology to patient diagnosis and treatment compared to western medicine (WM). Syndrome (ZHENG or pattern) is differentiated by a set of symptoms and signs examined from an individual by four main diagnostic methods: inspection, auscultation and olfaction, interrogation, and palpation which reflects the pathological and physiological changes of disease occurrence and development. Patient classification is to divide patients into several classes based on different criteria. In this paper, from the machine learning perspective, a survey on patient classification issue will be summarized on three major aspects of TCM: sign classification, syndrome differentiation, and disease classification. With the consideration of different diagnostic data analyzed by different computational methods, we present the overview for four subfields of TCM diagnosis, respectively. For each subfield, we design a rectangular reference list with applications in the horizontal direction and machine learning algorithms in the longitudinal direction. According to the current development of objective TCM diagnosis for patient classification, a discussion of the research issues around machine learning techniques with applications to TCM diagnosis is given to facilitate the further research for TCM patient classification.

A new type of combined ecological floating bed was developed on Swan Lake. It was combined with aquatic and hydrophilous plants and ancillary equipments that included floating controllers, water-cycling aerator systems and wave-making systems. Combined ecological floating bed changed the physical and chemical environment of Swan Lake and had an effect on the transportation and transformation of TN, NH4(+) -N and TP between overlying water and sediment during the experiment. The concentration change of TN, NH4(+) -N and TP and the effect of DO, Eh and pH on the nutrient in the overlying water and sediment were investigated. The results indicated that: in overlying water, the removal efficiencies of TN, NH4(+) -N and TP were 61.92%, 63.09% and 80.0%, respectively; the removal efficiencies of TN and NH4(+) -N were 23.79% and 37.04%, respectively; the concentration of TP increased by 43.71% in sediment during the experiment. The combined ecological floating bed influenced environmental factors such as DO, Eh and pH in overlying water in some degree. DO concentration rose from 8.7-8.9 mg x L(-1) to 9.3-10.4 mg x L(-1), and Eh concentration was from 163-178 mV to 191-198 mV. Both of them were higher than those in the contrast area. Phosphorus release was inhibited, and phosphorus adsorption was promoted through improving DO and Eh of the overlying water. The pH volatility was less and maintained between 7.51 and 8.32, and did not promote phosphorus release in sediment. TN, TP and NH4(+) -N in overlying water and TN and NH4(+) -N in sediment were significantly positively correlated with each other and with TP in sediment. The pH had no significant correlation with TN, TP and NH4(+) -N in overlying water and sediment. DO was significantly positively corrected with Eh in overlying water and was significantly negatively corrected with TP in sediment.

A clustering algorithm based selective neural networks ensemble (CLUSEN) is proposed to predict the degree of malignancy in brain glioma. Since the degree prediction of malignancy is critical before brain surgery, many learning methods are used like rule induction algorithm, single neural networks, support vector machines, etc. Ensemble learning methods can improve the generalization of single learning machine, and are becoming popular in the machine learning and medical data processing communities. The procedure of CLUSEN can efficiently remove redundancy learning individuals and help improve the diversity of ensemble methods. CLUSEN is used to predict the degree of malignancy in brain glioma. Experimental results on a set of brain glioma data show that, compared to support vector machines, rule induction and single neural networks, the classification accuracy of CLUSEN is higher.

Based on the double-injection effect, the dual-mode interference principle and the free-carrier plasma dispersion effect, a double-injection bifurcation optical active switch integrated on silicon-on-insulator has been proposed and fabricated. Its insertion loss and crosstalk are measured to be less than 5.32 and -14.4 dB, respectively, at wavelength 1.3 μm and the total switching current 90 mA. Response time is about 180 ns.

By means of CUSP model of catastrophe theory, this paper has studied the physics process of rockburst occurred on circular chamber. The present paper has not only described the instability process of rockburst more deeply, but also got the critical depth of plastic softening area of chamber that is valuable in the controlling engineering of rockburst. The chamber displacement jump and energy liberation have been derived. The influence of rock parameters on the rockburst has been discussed.

Based on the large cross-section single-mode rib waveguide condition, the total internal reflection and the free-carrier plasma dispersion effect, a silicon-on-insulator (SOI) 2 × 2 symmetric optical waveguide switch with a transverse injection structure has been proposed and fabricated, in which the SOI technique utilizes silicon and silicon dioxide thermal bonding and back-polishing. The device performance is measured at the wavelength of 1.3 μm. It shows that the crosstalk and insertion loss are less than - 18.1 and 4.8 dB, respectively, at an injection current of 60 mA, and response time is 110 ns.

To investigate the relation between various risk factors and benign prostatic hyperplasia (BPH).A population based case-control study was conducted, including 100 BPH patients over 60 years old living in suburb of Shenyang as study group, and 100 elderly men with non-BPH (excluding prostatic cancer and prostatitis) as control group. Chi(2) test and non-conditional logistic regression were used for monovariate analysis and multivariate analysis, respectively.Data from monovariate analysis showed that BPH incidence was significantly related to body weight index, cigarette smoking, alcohol drinking, meal intake at the beginning of 1980's, hypertention and prostatitis, respectively. While multivariate non-conditional logistic analysis showed that BPH was related to five factors: prostatitis (OR = 5.577, 95% CI: 2.147 - 14.482), monthly intake of meats at the beginning of 1980's (OR = 4.930, 95% CI: 2.404 - 10.111), diastolic blood pressure (OR = 1.050, 95% CI: 1.017 - 1.083), cigarette smoking (OR = 0.660, 95% CI: 0.500 - 0.872) and alcohol consumption (OR = 0.650, 95% CI: 0.480 - 0.881).Prostatitis, monthly excessive intake of meats at the beginning of 1980's and high diastolic blood pressure were possible risk factors for BPH, while heavy cigarette smoking and alcohol consumption were possible protective factors for BPH.