ABSTRACT
El llanto es una vía de comunicación del recién nacido con el medio circundante. Investigaciones acerca del llanto infantil han correlacionado características acústicas de éste con patologías, demostrándose que el llanto puede reflejar la integridad neurofisiológica del niño y dar una medida de su interacción con el ambiente y su desarrollo cognitivo y social. Esta contribución muestra cómo clasificar el llanto de neonatos con hipoxia y de un grupo de control, en normal o patológico, a través de una red neuronal artificial supervisada. Para implementar la red neuronal se aprovechan las posibilidades de la plataforma MATLAB®. El diseño y estructuración de la red considera algoritmo de aprendizaje o entrenamiento, iteraciones, pruebas e intervalos de clasificación, obteniéndose arquitectura y topología, y funcionalidades de la red neuronal que en la generalización proporciona la mejor clasificación. En el trabajo se aplica el método de selección de casos, el método acústico para extraer parámetros cuantitativos de la señal de llanto en tiempo, intensidad y frecuencia, así como métodos vinculados con el diseño, implementación y validación, con pruebas diagnósticas, de la red neuronal artificial obtenida para cumplir el objetivo del trabajo que es la generación de clases (clasificación del llanto). Con precisión del resultado de clasificación del 90 por ciento se está en condición de concebir una solución informática (agregando interfaz para interactuar con base de datos) para ayudar complementariamente al diagnóstico médico no invasivo usando el llanto del neonato provocado ante dolor(AU)
Cry from newborn (0-28 days) is a way of communication for the interaction with surrounding world. Infant cry researches provide information that correlate among cries acoustic features with pathologies. It has been demonstrated that the infant cry is able to reflect child neurophysiology integrity and give meaning from newborn interaction with environment, also cognitive and social development from child. This contribution shows how to classify the cry of neonates with hypoxia and of a control group, into normal or pathological, through a supervised artificial neural network. Network implementation makes use of MATLAB® platform possibilities. Design and structuring of network take into consideration aspects as training algorithm, iterations, tests and classification intervals. All these referred aspects give as result an architectural, topology and functionalities from neural network able to classify cry in generalization stage offering good outcome. Different methods are applied in this paper as selection of cases, acoustic methods in order to obtain quantitative parameters from cry signals (in time, intensity and frequency domain). Methods related with design, implementation and validation (diagnostic test) of an artificial neural network able to carry out the goal of this paper (classification of cry) are used. With accuracy results in cry classification about 90 percent, authors get ready conditions for an informatic solution (with addition of interface for data base interaction) for help as a non-invasive complement to medical diagnosis using cry from neonate induced by pain(AU)
Subject(s)
Humans , Male , Female , Infant, Newborn , Pain/etiology , Algorithms , Medical Informatics Applications , CryingABSTRACT
OBJECTIVE: To establish a method for predicting tablet hardness by near infrared diffuse reflection spectroscopy. METHODS: Tablet hardness value was obtained by hardness meter. Calibration model between NIR spectra and the hardness was establish by partial least squares regression (PLSR) method and BP-ANN method. RESULTS: Correlation coefficients (r), root mean squares error of cross-validation (RMSECV), and root mean square error of prediction (RMSEP) obtained by PLSR model were 0.977 8, 0.742 and 0.815 kg respectively. And the correlation coefficients of training set, monitor set and testing set by BP-ANN were 0.987 3, 0.985 6, and 0.986 8, with RSE% of 6.83%, 8.77%, and 6.69%, respectively. CONCLUSION: The prediction accuracy of BP-ANN nonlinear model is superior to the PLSR model.
ABSTRACT
<p><b>OBJECTIVE</b>To analyze the correlations between the incidence of tuberculosis and meteorological factors over the same period and previous periods including 1, 2 and 3 years ago, defined according to the Chinese medicine theory of five circuits (Wu Yun) and six qi, to establish medical-meteorological forecast models for the Beijing area of China.</p><p><b>METHODS</b>Data regarding the incidence of tuberculosis between 1990 and 2004 were obtained from the Beijing Center for Disease Control and Prevention, and the data regarding the meteorological factors (including daily average temperatures, wind speeds, precipitations, relative humidities, vapor pressures and low cloud covers) between 1987 and 2004 were collected from the Beijing Meteorological Observatory and analyzed. Descriptive statistics and a back-propagation artificial neural network were adopted to analyze the data.</p><p><b>RESULTS</b>There were significant correlations between the incidence of tuberculosis and the meteorological factors in the corresponding year and previous years. Among these correlations, wind speed was the factor with the strongest influence on tuberculosis (the standardized significance was 100%). Additionally, all prediction models would successfully established, suggesting the use of a collection of meteorological factors spanning from three years ago to the present is superior to the use of single data.</p><p><b>CONCLUSIONS</b>The incidence of tuberculosis in Beijing area is correlated to meteorological factors in the current year and previous years, which verifies the practicality of the theory of five circuits and six qi.</p>
Subject(s)
Humans , Beijing , Epidemiology , Forecasting , Medicine, Chinese Traditional , Meteorological Concepts , Tuberculosis , EpidemiologyABSTRACT
Objective To optimize the preparative procedure for stachydrine in Fructus Leonuri. Methods The preparation was screened by orthogonal experiment, and a mathematical model of relationship of extraction time, methanol concentration, and solid-liquid ratio with the content of stachydrine hydrochloride was established by using back-propagation (BP) neural network. And the process parameters were optimized with genetic algorithm (GA) . Results The optimum process parameters were as follows: extraction with 69% of methanol concentration and with solid-liquid ratio being 11 times for 62 min. The content of stachydrine obtained by BP neural network modeling and GA was higher than that achieved by orthogonal experiment. Conclusion The optimum preparative procedure could be achieved by combining BP modeling with GA. The model developed in this study was proved to be predictable and feasible for the optimization of process parameters of multi-dimension nonlinear system.
ABSTRACT
Objective: To correct the nonlinear error of sensor output, a new approach to sensor inverse modeling based on Back-Propagation Fuzzy Logical System (BP FS) is presented. Methods: The BP FS is a computationally efficient nonlinear universal approximator, which is capable of implementing complex nonlinear mapping from its input pattern space to the output with fast convergence speed. Results: The neuro-fuzzy hybrid system, i. e. BP FS, is then applied to construct nonlinear inverse model of pressure sensor. The experimental results show that the proposed inverse modeling method automatically compensates the associated nonlinear error in pressure estimation, and thus the performance of pressure sensor is significantly improved. Conclusion: The proposed method can be widely used in nonlinearity correction of various kinds of sensors to compensate the effects of nonlinearity and temperature on sensor output.
ABSTRACT
Objective To correct the nonlinear error of sensor output, a new approach to sensor inverse modeling based on Back-Propagation Fuzzy Logical System (BP FS) is presented. Methods The BP FS is a computationally efficient nonlinear universal approximator, which is capable of implementing complex nonlinear mapping from its input pattern space to the output with fast convergence speed. Results The neuro-fuzzy hybrid system, i.e. BP FS, is then applied to construct nonlinear inverse model of pressure sensor. The experimental results show that the proposed inverse modeling method automatically compensates the associated nonlinear error in pressure estimation, and thus the performance of pressure sensor is significantly improved. Conclusion The proposed method can be widely used in nonlinearity correction of various kinds of sensors to compensate the effects of nonlinearity and temperature on sensor output.
ABSTRACT
With the rapid growth of research and recognition about usefulness and importnace of the Nursing Diagnosis, the demand for application of Nursing Diagnosis has never been stronger. But in clinical field, not many nurses has used Nursing Diagnosis. Especially, nursing student have a difficulty to use Nursing Diagnosis because it demands for high level of capability of analyzing collected data and combining with relevant references. Therefore. this research has developed Nursing Diagnosis Self-learning Program using Back-propagating Neutral Network Model which is based on 98 surgery patients' data for nursing student. The twenty-six nursing diagnoses based on NANDA Taxonomy with 189 cases' reports and aid of 8 nursing experts wee determined to develop the program. To verify the usefulness of Nursing Diagnosis Self-learning Program constructed with the fully trained neural nets, the Program was tested with 70 real patients' data. The simulated output of program was compared with the judgement of the researcher and of two experts of nursing. The misdiagnosis rate of this program was eleven percent. This Program needs input of Signs and Symptoms, risk factors and 'related to' factors and also input the nursing diagnoses which a student selects. And than prints out two types of diagnoses. One is from the system and the other is what the student inputed. And the student makes the final diagnosis by refering the two types of diagnoses. Finally, the program prints out the completed diagnosis which problem combines with etiology in the diagnosis producing module. The program helps students to improve her capacity related to use Nursing Diagnosis.
Subject(s)
Humans , Classification , Diagnosis , Diagnostic Errors , Neural Networks, Computer , Nursing Diagnosis , Nursing , Risk Factors , Students, NursingABSTRACT
This study was performed to find out nutritional factors associated with identification of chronic disease using neural network model. A dietary survey with 24-hour recall method was conducted together with a health survey including health qustionnaire, physical examination and glucose tolerance test to 2037 adults over 30 years of age in rural area of Korea. Subjects have been classified into groups with and without disease depending on each disease criteria(groups with desease include those who are newly diagnosed to have diabetes, hypertension, hyperlipidemia and obesity, respictively). Neural network method has been applied to predict disease using selected 3 nutrients out of 12 nutrient elements as input information and resulting outputs which designate either disease or without disease group. Backpropagation learning algorithm has been applied to train neural network structure assigning weight factors connecting each node. 20 subjects from both disease and without disease group have been collected to train neural network structure and remaining subjects were later used in order to test the validity of the trained structure. In order to quantify contrivution ratio of each nutrients for predictiong disease status, number of appearance frequency of each nutrients in the top 20 prediction rate has been compared. When a nutrient used 10 times for prediction, its appearance rate was calculated to be 50%. vitamin C, vitamin A and iron showed appearance rate of 70%, 40% and 30%, respectively for predicting diavetes. vitain C, Fat and beta-carotene showed appearance rate of 60%, 60% and 40% for predicting hypertension. For predicting hyperlipidemia, appearance rate of vitamin C, iron and energy was 65%, 55% and 30%, respectively. From the study, vitamin C was shown to be prominent in predicting chronic disease groups from subjects.