Search

Objectives: The accuracy of ultrasound (US) measurements is operator dependent. In order to decrease the operator-dependent errors in estimated fetal weight (EFW), a model selection analysis was undertaken to select significant compensation weighting factors on ultrasonographic parameters to support artificial neural network (ANN), and thus to enhance the accuracy of fetal weight estimation. Materials and Methods: In total, 2127 singletons were examined by prenatal US within 3 days before delivery for ANN development, and another 100 cases were selected from new operators for evaluation. First, correlation analysis was used to analyze the differences between the prenatal and postnatal parameters. Second, Akaike information criterion (AIC) was used to determine the number of database partition and optimal weightings for compensating the input parameters of the ANN model. Finally, minimum mean squared error (MMSE) mode was utilized to determine the optimal EFW. Results: EFW of the proposed compensation model using AIC and MMSE showed mean absolute percent error of 5.1 ± 3.1% and mean absolute error of 158.9 ± 96.2 g. When comparing the accuracy of EFW, our model using AIC and MMSE was superior to those conventional EFW formulas (all p

Objectives: A novel analysis was undertaken to select a significant ultrasonographic parameter (USP) for classifying fetuses to support artificial neural network (ANN), and thus to enhance the accuracy of fetal weight estimation. Methods: In total, 2127 singletons were examined by prenatal ultrasound within 3 days before delivery. First, correlation analysis was used to determine a significant USP for fetal grouping. Second, K-means algorithm was utilized for fetal size classification based on the selected USP. Finally, stepwise regression analysis was used to examine input parameters of the ANN model. Results: The estimated fetal weight (EFW) of the new model showed mean absolute percent error (MAPE) of 5.26 ± 4.14% and mean absolute error (MAE) of 157.91 ± 119.90 g. Comparison of EFW accuracy showed that the new model significantly outperformed the commonly-used EFW formulas (all p < 0.05). Conclusion: We proved the importance of choosing a specific grouping parameter for ANN to improve EFW accuracy.

An essential task in a genomic analysis of a human disease is limiting the number of strongly associated genes when studying susceptibility to the disease. The goal of this study was to compare computational tools with and without feature selection for predicting osteoporosis outcome in Taiwanese women based on genetic factors such as single nucleotide polymorphisms (SNPs). To elucidate relationships between osteoporosis and SNPs in this population, three classification algorithms were applied: multilayer feedforward neural network (MFNN), naive Bayes, and logistic regression. A wrapper-based feature selection method was also used to identify a subset of major SNPs. Experimental results showed that the MFNN model with the wrapper-based approach was the best predictive model for inferring disease susceptibility based on the complex relationship between osteoporosis and SNPs in Taiwanese women. The findings suggest that patients and doctors can use the proposed tool to enhance decision making based on clinical factors such as SNP genotyping data.

BACKGROUND: An adaptive-network-based fuzzy inference system (ANFIS) was compared with an artificial neural network (ANN) in terms of accuracy in predicting the combined effects of temperature (10.5 to 24.5°C), pH level (5.5 to 7.5), sodium chloride level (0.25% to 6.25%) and sodium nitrite level (0 to 200 ppm) on the growth rate of Leuconostoc mesenteroides under aerobic and anaerobic conditions. METHODS: THE ANFIS AND ANN MODELS WERE COMPARED IN TERMS OF SIX STATISTICAL INDICES CALCULATED BY COMPARING THEIR PREDICTION RESULTS WITH ACTUAL DATA: mean absolute percentage error (MAPE), root mean square error (RMSE), standard error of prediction percentage (SEP), bias factor (Bf), accuracy factor (Af), and absolute fraction of variance (R (2)). Graphical plots were also used for model comparison. CONCLUSIONS: The learning-based systems obtained encouraging prediction results. Sensitivity analyses of the four environmental factors showed that temperature and, to a lesser extent, NaCl had the most influence on accuracy in predicting the growth rate of Leuconostoc mesenteroides under aerobic and anaerobic conditions. The observed effectiveness of ANFIS for modeling microbial kinetic parameters confirms its potential use as a supplemental tool in predictive mycology. Comparisons between growth rates predicted by ANFIS and actual experimental data also confirmed the high accuracy of the Gaussian membership function in ANFIS. Comparisons of the six statistical indices under both aerobic and anaerobic conditions also showed that the ANFIS model was better than all ANN models in predicting the four kinetic parameters. Therefore, the ANFIS model is a valuable tool for quickly predicting the growth rate of Leuconostoc mesenteroides under aerobic and anaerobic conditions.

In addition to fierce intra-industry competition, chain businesses face many external problems. One problem is that the traditional stand-alone point-of-sale (POS) system has been unable to quickly grasp the operating performance and management of each branch. Another problem is that equipment costs continue to increase with the number of stores. Therefore, the Taiwan government has included cloud computing in its initiative to promote emerging smart industries. In addition, the world is actively investing in the cloud computing industry. Owing to security concerns, however, many domestic enterprises are reluctant to adopt cloudbased services. Secure cloud service applications must be developed to maintain and expand market share. We implement a Cloud Web POS 365 Platform with a multifactor and threshold authentication. Then, we use the required and threshold parameters to generate a key for encryption/decryption. Finally, we deconstruct a ciphertext to partial ciphertext segments and store it in each partial ciphertext database. Hence, this study contains 4 security levels. First, users need the required parameters including positioning information; when the machine is not in the set range, it cannot be used. Second, users need the threshold parameters stored in a radio frequency identification card and obtain enough card information to log into the system. Those parameters above are generating the encryption/decryption key. Third, we use a symmetric encryption module to encrypt data. Fourth, we design a deconstruction/reconstruction module to store partial ciphertext segments separately. Therefore, the proposed scheme is more secure than other cloud web POS systems. Moreover, its web-based architecture enables easy implementation in any platform. As a result, the proposed scheme would be effective for increasing the operating efficiency of chain businesses. [ABSTRACT FROM AUTHOR]

Motivated by the need for better control of standards of a certification examination for interpreters in Taiwan, this exploratory study aimed at identifying indicators that may be used to predict source material difficulty for consecutive interpreting. A combination of quantifiable measures — readability level, information density and new concept density — was used to examine different aspects of three English source materials. Expert judgment was also used as a more holistic method of judging source material difficulty. The results of these analyses were compared with two groups of student interpreters’ performance on consecutive interpreting of the source materials into Mandarin Chinese. The participants’ assessment of speech difficulty after the interpreting task was also compared with the other measures and the expert judgment. The quantifiable measures all failed statistically in predicting source material difficulty, possibly due to the very small sample size of the materials or to the fact that the materials were very similar in the aspects assessed by these measures. A trend emerged to suggest that information density and sentence length may be potentially useful indicators for predicting source material difficulty. It was also shown that source material difficulty affected the performance of lower-skilled interpreters more than that of higher-skilled interpreters. [ABSTRACT FROM AUTHOR]

This paper proposes an approach to segmenting and identifying mixed-language speech. A delta Bayesian information criterion (delta-BIC) is firstly applied to segment the input speech utterance into a sequence of language-dependent segments using acoustic features. A VQ-based bi-gram model is used to characterize the acoustic-phonetic dynamics of two consecutive codewords in a language. Accordingly the language-specific acoustic-phonetic property of sequence of phones was integrated in the identification process. A Gaussian mixture model (GMM) is used to model codeword occurrence vectors orthonormally transformed using latent semantic analysis (LSA) for each language. dependent segment. A filtering method is used to smooth the hypothesized language sequence and thus eliminate noise-like components of the detected language sequence generated by the maximum likelihood estimation. Finally, a dynamic programming method is used to determine globally the language boundaries. Experimental results show that for Mandarin, English, and Taiwanese, a recall rate of 0.87 for language boundary segmentation was obtained. Based on this recall rate, the proposed approach achieved language identification accuracies of 92.1% and 74.9% for single-language and mixed-language speech, respectively. [ABSTRACT FROM AUTHOR]

This paper proposes a novel approach to the generation of Chinese sentences from ill-formed Taiwanese Sign Language (TSL) for people with hearing impairments. First, a sign icon-based virtual keyboard is constructed to provide a visualized interface to retrieve sign icons from a sign database. A proposed language model (LM), based on a predictive sentence template (PST) tree, integrates a statistical variable n-gram LM and linguistic constraints to deal with the translation problem from ill-formed sign sequences to grammatical written sentences. The PST tree trained by a corpus collected from the deaf schools was used to model the correspondence between signed and written Chinese. In addition, a set of phrase formation rules, based on trigger pair category, was derived for sentence pattern expansion. These approaches improved the efficiency of text generation and the accuracy of word prediction and, therefore, improved the input rate. For the assessment of practical communication aids, a reading-comprehension training program with ten profoundly deaf students was undertaken in a deaf school in Tainan, Taiwan. Evaluation results show that the literacy aptitude test and subjective satisfactory level are significantly improved. [ABSTRACT FROM AUTHOR]