ABSTRACT
The ambulatory assessment of vocal function can be significantly enhanced by having access to physiologically based features that describe underlying pathophysiological mechanisms in individuals with voice disorders. This type of enhancement can improve methods for the prevention, diagnosis, and treatment of behaviorally based voice disorders. Unfortunately, the direct measurement of important vocal features such as subglottal pressure, vocal fold collision pressure, and laryngeal muscle activation is impractical in laboratory and ambulatory settings. In this study, we introduce a method to estimate these features during phonation from a neck-surface vibration signal through a framework that integrates a physiologically relevant model of voice production and machine learning tools. The signal from a neck-surface accelerometer is first processed using subglottal impedance-based inverse filtering to yield an estimate of the unsteady glottal airflow. Seven aerodynamic and acoustic features are extracted from the neck surface accelerometer and an optional microphone signal. A neural network architecture is selected to provide a mapping between the seven input features and subglottal pressure, vocal fold collision pressure, and cricothyroid and thyroarytenoid muscle activation. This non-linear mapping is trained solely with 13,000 Monte Carlo simulations of a voice production model that utilizes a symmetric triangular body-cover model of the vocal folds. The performance of the method was compared against laboratory data from synchronous recordings of oral airflow, intraoral pressure, microphone, and neck-surface vibration in 79 vocally healthy female participants uttering consecutive /pæ/ syllable strings at comfortable, loud, and soft levels. The mean absolute error and root-mean-square error for estimating the mean subglottal pressure were 191 Pa (1.95 cm H2O) and 243 Pa (2.48 cm H2O), respectively, which are comparable with previous studies but with the key advantage of not requiring subject-specific training and yielding more output measures. The validation of vocal fold collision pressure and laryngeal muscle activation was performed with synthetic values as reference. These initial results provide valuable insight for further vocal fold model refinement and constitute a proof of concept that the proposed machine learning method is a feasible option for providing physiologically relevant measures for laboratory and ambulatory assessment of vocal function.
ABSTRACT
En la primera parte de este trabajo se presenta la construcción y el desarrollo de la Prueba Argentina Psicolingüística de Denominación de Imágenes (PAPDI). La misma tiene dos características que la diferencian de las pruebas actualmente disponibles en nuestro medio: (a) Los estímulos fueron seleccionados controlando las variables que afectan las etapas de análisis visual y reconocimiento de la imagen (Complejidad visual y Concordancia con la imagen) y el conocimiento conceptual (Variabilidad de la Imagen y Familiaridad) del dibujo del objeto también se manipularon las variables (Frecuencia léxica y Edad de adquisición) que afectan la etapa que busca evaluar la prueba: activación, selección y recuperación de la etiqueta léxica. Los índices que operacionalizan a las variables son dependientes de la cultura y de la lengua de dónde se obtengan, por ello se tomaron sus valores de normas argentinas. (b) Las claves semánticas han sido elaboradas cuidadosamente, siguiendo las normas de producción de atributos semánticos también recolectadas en nuestro país, ya que el uso de las mismas permite generar hipótesis más precisas acerca de la localización del déficit. Por lo tanto, se presenta el proceso de construcción y desarrollo de la prueba mostrando cómo se tuvieron en cuenta estas particularidades, cuáles variables han sido controladas y cuáles manipuladas, y su justificación. Por último, los resultados de la prueba piloto permitieron seleccionar 30 ítems y la constitución de la versión final de la prueba, que es lo suficientemente breve para su uso en la clínica y de libre acceso.
In the current paper, first part, we present the design and development of the Argentinean Psycholinguistic Image Naming Test (PAPDI). This test has two characteristics that make it different from those currently available in our country: (a) the stimuli have been selected by controlling the variables that affect the stages of visual analysis and image recognition (Visual Complexity and Image Agreement) and conceptual knowledge (Image Variability and Familiarity) of the drawing of the object; and the variables (Frequency of Use and Age of Acquisition) that affect the stage that seeks to evaluate the test were manipulated: activation, selection and retrieval of the lexical label (phonological form of the name), to generate a gradient of difficulty in the items of the test. However, the variable Naming Agreement, that also affects this stage, was controlled to selected pictures with only one predominating name. The quantification of these factors by means of specific variables is influenced by the cultural and linguistic context from which they were obtained, consequently we took their values from Argentinean normative data base and (b) semantic cues were elaborated according to semantic feature production norms also from Argentina. The phonological cues corresponded to the first syllable of the object's name. Their use allows to generate more precise hypotheses about the location of the deficit and allows to infer relevant information about the cognitive profile. Therefore, the process of construction and development of the test is presented, showing how these particularities were taken into consideration, which variables were controlled and which were manipulated, and their justification. For the design of the test 62, black and white pictures were originally selected from the 400 pictures taken from Cycowicz's set (of frequent use in Experimental Psychology and Neuropsychology). They correspond to concrete concepts from different semantic categories belonging to both living andnon living domains. The criteria used to include those images were that they were moderately complex (in quantity of lines and details), moderately familiar and that they had a mean image variability value. They had to have a medium to high concordance degree between the mental image and its corresponding representation. Besides, these images had to have a naming agreement superior to 80%. To conform the first version of the test, the 62 images selected were ordered according to the variable Age of Acquisition and each image was assigned its corresponding semantic and phonological cues. The former were extracted from the Argentinean norms considering the most relevant features. The second consisted on the word´s first syllable. A pilot study with healthy population (n = 50), of different ages, both sexes and three educational levels, was carried out through which 30 images were selected that constituted the test´s final version. The deleted items were: figures that did not evokea univocal response, that is, they were easily confused with other objects; figures that had more than one acceptable answer although not all were correct; and items that presented a ceiling effect where the scores accumulated in high scores and did not allow to discriminate the participants' ability to select, retrieve and produce a word (only 25 and 30% of the items were retained). In the test´s final version, the items were reordered according to the Age of Acquisition and Frequency values and their difficulty in naming (following the percentage of successes in the spontaneous response). And images that were contain in another naming tests were deleted. The test is short enough for use in the clinic and freely accessible. The second part of the paper presents the psychometric studies that provide evidence of their relevance and validity.
ABSTRACT
Este trabalho teve por objetivo estimar e validar um modelo de simulação para a cultura da soja na região de Santiago, RS. Para isso, foi necessário desenvolver procedimento experimental, para geração de dados necessários a fim de calibrar e testar o modelo de simulação, da extração da água do solo, e a respectiva resposta em produção da cultura. Como resultados experimentais, obtiveram-se variações no crescimento e desenvolvimento da cultura da soja com a aplicação de diferentes estratégias de irrigação, resultando em produtividades distintas. O aumento nas lâminas de irrigação proporcionou incremento nos componentes de produção. Os testes dos dados simulados demonstraram variação inferior a 8% na comparação dos dados medidos, no experimento de campo, com os dados simulados pelo modelo. O modelo de produção proposto foi capaz de simular satisfatoriamente o acúmulo de matéria seca total e a produção de grãos para a cultura da soja.(AU)
This study aimed estimate and validate a simulation model for the soybean crop in the region of Santiago, RS, Brazil which will, to the agricultural sector, support in decision making. For this, it was necessary to develop experimental procedure for generating the data needed to calibrate and test the simulation model, the extraction of water from the soil, and the response in crop production. As experimental results, obtained variations in the growth and development of soybean by applying different strategies for irrigation, resulting in different productivities. The increase in irrigation provided an increment in production components. Tests of simulated data showed a variation of less than 8% when comparing the data measured in field experiments with simulated data. The production model proposed was successfully able to simulate the accumulation of total dry matter and grain production for soybeans.(AU)
Subject(s)
Glycine max , Agricultural IrrigationABSTRACT
Este trabalho teve por objetivo estimar e validar um modelo de simulação para a cultura da soja na região de Santiago, RS. Para isso, foi necessário desenvolver procedimento experimental, para geração de dados necessários a fim de calibrar e testar o modelo de simulação, da extração da água do solo, e a respectiva resposta em produção da cultura. Como resultados experimentais, obtiveram-se variações no crescimento e desenvolvimento da cultura da soja com a aplicação de diferentes estratégias de irrigação, resultando em produtividades distintas. O aumento nas lâminas de irrigação proporcionou incremento nos componentes de produção. Os testes dos dados simulados demonstraram variação inferior a 8% na comparação dos dados medidos, no experimento de campo, com os dados simulados pelo modelo. O modelo de produção proposto foi capaz de simular satisfatoriamente o acúmulo de matéria seca total e a produção de grãos para a cultura da soja.
This study aimed estimate and validate a simulation model for the soybean crop in the region of Santiago, RS, Brazil which will, to the agricultural sector, support in decision making. For this, it was necessary to develop experimental procedure for generating the data needed to calibrate and test the simulation model, the extraction of water from the soil, and the response in crop production. As experimental results, obtained variations in the growth and development of soybean by applying different strategies for irrigation, resulting in different productivities. The increase in irrigation provided an increment in production components. Tests of simulated data showed a variation of less than 8% when comparing the data measured in field experiments with simulated data. The production model proposed was successfully able to simulate the accumulation of total dry matter and grain production for soybeans.