scholarly journals A Deep Neural Network Based Glottal Flow Model for Predicting Fluid-Structure Interactions during Voice Production

2020 ◽  
Vol 10 (2) ◽  
pp. 705
Author(s):  
Yang Zhang ◽  
Xudong Zheng ◽  
Qian Xue
Keyword(s):  
Neural Network ◽  
Flow Resistance ◽  
Deep Neural Network ◽  
Shape Function ◽  
Resistance Coefficient ◽  
Prediction Performance ◽  
Fluid Structure Interactions ◽  
Fluid Structure ◽  
Voice Production ◽  
Glottal Flow

This paper proposes a machine-learning based reduced-order model that can provide fast and accurate prediction of the glottal flow during voice production. The model is based on the Bernoulli equation with a viscous loss term predicted by a deep neural network (DNN) model. The training data of the DNN model is a Navier-Stokes (N-S) equation-based three-dimensional simulation of glottal flows in various glottal shapes generated by a synthetic shape function, which can be obtained by superimposing the instantaneous modal displacements during vibration on the prephonatory geometry of the glottal shape. The input parameters of the DNN model are the geometric and flow parameters extracted from discretized cross sections of the glottal shapes and the output target is the corresponding flow resistance coefficient. With this trained DNN-Bernoulli model, the flow resistance coefficient as well as the flow rate and pressure distribution in any given glottal shape generated by the synthetic shape function can be predicted. The model is further coupled with a finite-element method based solid dynamics solver for simulating fluid-structure interactions (FSI). The prediction performance of the model for both static shape and FSI simulations is evaluated by comparing the solutions to those obtained by the Bernoulli and N-S model. The model shows a good prediction performance in accuracy and efficiency, suggesting a promise for future clinical use.

scholarly journals Effect of Subglottic Stenosis on Vocal Fold Vibration and Voice Production Using Fluid–Structure–Acoustics Interaction Simulation

2021 ◽  
Vol 11 (3) ◽  
pp. 1221
Author(s):  
Dariush Bodaghi ◽  
Qian Xue ◽  
Xudong Zheng ◽  
Scott Thomson
Keyword(s):  
Flow Rate ◽  
Vocal Fold ◽  
Flow Resistance ◽  
Signal To Noise Ratio ◽  
Area Ratio ◽  
Subglottic Stenosis ◽  
Fluid Structure ◽  
Voice Production ◽  
Vocal Fold Vibration ◽  
Glottal Flow

An in-house 3D fluid–structure–acoustic interaction numerical solver was employed to investigate the effect of subglottic stenosis (SGS) on dynamics of glottal flow, vocal fold vibration and acoustics during voice production. The investigation focused on two SGS properties, including severity defined as the percentage of area reduction and location. The results show that SGS affects voice production only when its severity is beyond a threshold, which is at 75% for the glottal flow rate and acoustics, and at 90% for the vocal fold vibrations. Beyond the threshold, the flow rate, vocal fold vibration amplitude and vocal efficiency decrease rapidly with SGS severity, while the skewness quotient, vibration frequency, signal-to-noise ratio and vocal intensity decrease slightly, and the open quotient increases slightly. Changing the location of SGS shows no effect on the dynamics. Further analysis reveals that the effect of SGS on the dynamics is primarily due to its effect on the flow resistance in the entire airway, which is found to be related to the area ratio of glottis to SGS. Below the SGS severity of 75%, which corresponds to an area ratio of glottis to SGS of 0.1, changing the SGS severity only causes very small changes in the area ratio; therefore, its effect on the flow resistance and dynamics is very small. Beyond the SGS severity of 75%, increasing the SGS severity, leads to rapid increases of the area ratio, resulting in rapid changes in the flow resistance and dynamics.

scholarly journals Deep Neural Network and Monte Carlo Tree Search applied to Fluid-Structure Topology Optimization

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Audrey Gaymann ◽  
Francesco Montomoli
Keyword(s):  
Neural Network ◽  
Monte Carlo ◽  
Topology Optimization ◽  
Deep Neural Network ◽  
Level Set Methods ◽  
Test Case ◽  
Tree Search ◽  
Fluid Structure ◽  
Monte Carlo Tree Search ◽  
Structure Topology

Abstract This paper shows the application of Deep Neural Network algorithms for Fluid-Structure Topology Optimization. The strategy offered is a new concept which can be added to the current process used to study Topology Optimization with Cellular Automata, Adjoint and Level-Set methods. The design space is described by a computational grid where every cell can be in two states: fluid or solid. The system does not require human intervention and learns through an algorithm based on Deep Neural Network and Monte Carlo Tree Search. In this work the objective function for the optimization is an incompressible fluid solver but the overall optimization process is independent from the solver. The test case used is a standard duct with back facing step where the optimizer aims at minimizing the pressure losses between inlet and outlet. The results obtained with the proposed approach are compared to the solution via a classical adjoint topology optimization code.

scholarly journals Development of a deep neural network for predicting 6-hour average PM<sub>2.5</sub> concentrations up to two subsequent days using various training data

2021 ◽  
Author(s):  
Jeong-Beom Lee ◽  
Jae-Bum Lee ◽  
Youn-Seo Koo ◽  
Hee-Yong Kwon ◽  
Min-Hyeok Choi ◽  
...  
Keyword(s):  
Neural Network ◽  
Air Quality ◽  
Deep Neural Network ◽  
Model Performance ◽  
Prediction Performance ◽  
Training Data ◽  
Forecasting Model ◽  
Observation Data ◽  
Forecast Data ◽  
Air Quality Forecasting

Abstract. This study aims to develop a deep neural network (DNN) model as an artificial neural network (ANN) for the prediction of 6-hour average fine particulate matter (PM2.5) concentrations for a three-day period—the day of prediction (D+0), one day after prediction (D+1) and two days after prediction (D+2)—using observation data and forecast data obtained via numerical models. The performance of the DNN model was comparatively evaluated against that of the currently operational Community Multiscale Air Quality (CMAQ) modelling system for air quality forecasting in South Korea. In addition, the effect on predictive performance of the DNN model on using different training data was analyzed. For the D+0 forecast, the DNN model performance was superior to that of the CMAQ model, and there was no significant dependence on the training data. For the D+1 and D+2 forecasts, the DNN model that used the observation and forecast data (DNN-ALL) outperformed the CMAQ model. The root-mean-squared error (RMSE) of DNN-ALL was lower than that of the CMAQ model by 2.2 μgm−3, and 3.0 μgm−3 for the D+1 and D+2 forecasts, respectively, because the overprediction of higher concentrations was curtailed. An IOA increase of 0.46 for D+1 prediction and 0.59 for the D+2 prediction was observed in case of the DNN-ALL model compared to the IOA of the DNN model that used only observation data (DNN-OBS). In additionally, An RMSE decrease of 7.2 μgm−3 for the D+1 prediction and 6.3 μgm−3 for the D+2 prediction was observed in case of the DNN-ALL model, compared to the RMSE of DNN-OBS, indicating that the inclusion of forecast data in the training data greatly affected the DNN model performance. Considering the prediction of the 6-hour average PM2.5 concentration, the 8.8 μgm−3 RMSE of the DNN-ALL model was 2.7 μgm−3 lower than that of the CMAQ model, indicating the superior prediction performance of the former. These results suggest that the DNN model could be utilized as a better-performing air quality forecasting model than the CMAQ, and that observation data plays an important role in determining the prediction performance of the DNN model for D+0 forecasting, while prediction data does the same for D+1 and D+2 forecasting. The use of the proposed DNN model as a forecasting model may result in a reduction in the economic losses caused by pollution-mitigation policies and aid better protection of public health.

scholarly journals Experimental study of overland flow resistance coefficient model of grassland based on BP neural network

2018 ◽  
Vol 38 ◽  
pp. 01030 ◽  
Author(s):  
Peng Jiao ◽  
Er Yang ◽  
Yong Xin Ni
Keyword(s):  
Neural Network ◽  
Bp Neural Network ◽  
Prediction Error ◽  
Flow Resistance ◽  
Rainfall Intensity ◽  
Overland Flow ◽  
Resistance Coefficient ◽  
Simulation Accuracy ◽  
Average Prediction Error ◽  
Optimized Model

The overland flow resistance on grassland slope of 20° was studied by using simulated rainfall experiments. Model of overland flow resistance coefficient was established based on BP neural network. The input variations of model were rainfall intensity, flow velocity, water depth, and roughness of slope surface, and the output variations was overland flow resistance coefficient. Model was optimized by Genetic Algorithm. The results show that the model can be used to calculate overland flow resistance coefficient, and has high simulation accuracy. The average prediction error of the optimized model of test set is 8.02%, and the maximum prediction error was 18.34%.

Effect of Supraglottal Acoustics On Fluid-Structure Interaction During Human Voice Production

2021 ◽  
Author(s):  
Dariush Bodaghi ◽  
Weili Jiang ◽  
Qian Xue ◽  
Xudong Zheng
Keyword(s):  
Flow Rate ◽  
Vocal Fold ◽  
Splitting Method ◽  
Acoustic Resonance ◽  
Coupled Processes ◽  
Fluid Structure Interactions ◽  
Fluid Structure ◽  
Voice Production ◽  
Acoustic Perturbation ◽  
Human Voice

Abstract A hydrodynamic/acoustic splitting method was used to examine the effect of supraglottal acoustics on fluid-structure interactions during human voice production in a two-dimensional computational model. The accuracy of the method in simulating compressible flows in typical human airway conditions was verified by comparing it to full compressible flow simulations. The method was coupled with a three-mass model of vocal fold lateral motion to simulate fluid-structure interactions during human voice production. By separating the acoustic perturbation components of the airflow, the method allows isolation of the role of supraglottal acoustics in fluid-structure interactions. The results showed that an acoustic resonance between a higher harmonic of the sound source and the first formant of the supraglottal tract occurred during normal human phonation when the fundamental frequency was much lower than the formants. The resonance resulted in acoustic pressure perturbation at the glottis which was of the same order as the incompressible flow pressure and found to affect vocal fold vibrations and glottal flow rate waveform. Specifically, the acoustic perturbation delayed the opening of the glottis, reduced the vertical phase difference of vocal fold vibrations, decreased flow rate and maximum flow deceleration rate at the glottal exit; yet, they had little effect on glottal opening. The results imply that the sound generation in the glottis and acoustic resonance in the supraglottal tract are coupled processes during human voice production and computer modeling of vocal fold vibrations needs to include supraglottal acoustics for accurate predictions.

A Physical Model of the Vocal Folds

2003 ◽  
Author(s):  
Scott L. Thomson ◽  
Luc Mongeau ◽  
Steven H. Frankel
Keyword(s):  
Physical Model ◽  
Vocal Fold ◽  
Vocal Folds ◽  
Operating Conditions ◽  
Physical Models ◽  
Length Scale ◽  
Fluid Structure Interactions ◽  
Fluid Structure ◽  
Voice Production ◽  
Flexible Polymer

Voice production is a result of the nonlinear, coupled interaction between laryngeal airflow and vocal fold tissue dynamics. Studying these fluid-structure interactions can contribute to the understanding of the mechanisms of speech production, leading to improved surgical, clinical, and pedagogical care. Aside from experiments using excised larynges (e.g., Berry et al., 2001) and a model of the superficial vocal fold layer (e.g., Chan et al., 1997), no studies appear to have been reported in which self-oscillating physical models were used that were similar to the human vocal folds in the following aspects: length scale, geometry, and dynamic and mechanical behavior. This paper describes a self-oscillating physical model designed to more closely represent the human vocal folds in terms of the above key parameters. The model was constructed using a flexible polymer casting and exhibited regular, self-sustained, large-amplitude oscillations at frequencies and operating conditions close to those found in human phonation. The model demonstrated potential for further studies involving laryngeal fluid-structure interactions.

scholarly journals A Novel Hybrid Deep Neural Network Model to Predict the Refrigerant Charge Amount of Heat Pumps

2020 ◽  
Vol 12 (7) ◽  
pp. 2914
Author(s):  
Jun Kwon Hwang ◽  
Patrick Nzivugira Duhirwe ◽  
Geun Young Yun ◽  
Sukho Lee ◽  
Hyeongjoon Seo ◽  
...  
Keyword(s):  
Neural Network ◽  
Thermodynamic Model ◽  
Deep Neural Network ◽  
Heat Pumps ◽  
Current Data ◽  
Prediction Performance ◽  
Data Driven ◽  
Electric Heat ◽  
Model Training ◽  
Optimum Charge

Improper refrigerant charge amount (RCA) is a recurring fault in electric heat pump (EHP) systems. Because EHP systems show their best performance at optimum charge, predicting the RCA is important. There has been considerable development of data-driven techniques for predicting RCA; however, the current data-driven approaches for estimating RCA suffer from poor generalization and overfitting. This study presents a hybrid deep neural network (DNN) model that combines both a basic DNN model and a thermodynamic model to counter the abovementioned challenges of existing data-driven approaches. The data for designing models were collected from two EHP systems with different specifications, which were used for the training and testing of models. In addition to the data obtained using the basic DNN model, the hybrid DNN model uses the thermodynamic properties as a thermodynamic model. The testing results show that the hybrid DNN model has a prediction performance of 93%, which is 21% higher than that of the basic DNN model. Furthermore, for model training and model testing, the hybrid DNN model has a 6% prediction performance difference, indicating its reliable generalization capabilities. To summarize, the hybrid DNN model improves data-driven approaches and can be used for designing efficient and energy-saving EHP systems.

scholarly journals Alpha-Frag: a deep neural network for fragment presence prediction improves peptide identification by data independent acquisition mass spectrometry

2021 ◽  
Author(s):  
Jian Song ◽  
Fangfei Zhang ◽  
Changbin Yu
Keyword(s):  
Neural Network ◽  
Mass Spectrometry ◽  
Deep Neural Network ◽  
Peptide Identification ◽  
Prediction Performance ◽  
Fragmentation Mechanism ◽  
Statistical Validation ◽  
Test Dataset ◽  
Data Independent Acquisition ◽  
Fragment Ions

ABSTRACTMotivationIdentification of peptides in data-independent acquisition (DIA) mass spectrometry (MS) typically relies on the scoring for the peak groups upon extracted chromatograms of fragment ions. Expanding fragment scoring features closer to the genuine experimental spectra can improve DIA identification. Deep learning is able to predict fragment presence without understanding the fragmentation mechanism that can enrich the scoring features in DIA identification.ResultsIn this work, we developed a deep neural network-based model, Alpha-Frag, to predict the fragment ions that should be present for a given peptide by reporting their probabilities of existence. The prediction performance was evaluated in terms of intersection over union (IoU), and Alpha-Frag achieved an average of >0.7 and outperformed substantially the benchmarks across the validation datasets. Furthermore, qualitative scores based on Alpha-Frag were designed and incorporated into the peptide statistical validation tools as auxiliary scores. Our preliminary experiments show that the qualitative scores by Alpha-Frag are profitable for DIA identification, especially in the case of short gradient, and yielded an increase of 10.1%-29.3% improvements for the test dataset compared to the same scoring strategy but using Prosit.Availability and ImplementationSource code and the trained model are available at www.github.com/YuAirLab/Alpha-Frag.

scholarly journals Network-based drug sensitivity prediction

2020 ◽  
Vol 13 (S11) ◽  
Author(s):  
Khandakar Tanvir Ahmed ◽  
Sunho Park ◽  
Qibing Jiang ◽  
Yunku Yeu ◽  
TaeHyun Hwang ◽  
...  
Keyword(s):  
Neural Network ◽  
Feature Selection ◽  
Drug Response ◽  
Deep Neural Network ◽  
Drug Sensitivity ◽  
Feature Selection Method ◽  
Network Models ◽  
Prediction Performance ◽  
Neural Network Models ◽  
Drug Sensitivity Prediction

Abstract Background Drug sensitivity prediction and drug responsive biomarker selection on high-throughput genomic data is a critical step in drug discovery. Many computational methods have been developed to serve this purpose including several deep neural network models. However, the modular relations among genomic features have been largely ignored in these methods. To overcome this limitation, the role of the gene co-expression network on drug sensitivity prediction is investigated in this study. Methods In this paper, we first introduce a network-based method to identify representative features for drug response prediction by using the gene co-expression network. Then, two graph-based neural network models are proposed and both models integrate gene network information directly into neural network for outcome prediction. Next, we present a large-scale comparative study among the proposed network-based methods, canonical prediction algorithms (i.e., Elastic Net, Random Forest, Partial Least Squares Regression, and Support Vector Regression), and deep neural network models for drug sensitivity prediction. All the source code and processed datasets in this study are available at https://github.com/compbiolabucf/drug-sensitivity-prediction. Results In the comparison of different feature selection methods and prediction methods on a non-small cell lung cancer (NSCLC) cell line RNA-seq gene expression dataset with 50 different drug treatments, we found that (1) the network-based feature selection method improves the prediction performance compared to Pearson correlation coefficients; (2) Random Forest outperforms all the other canonical prediction algorithms and deep neural network models; (3) the proposed graph-based neural network models show better prediction performance compared to deep neural network model; (4) the prediction performance is drug dependent and it may relate to the drug’s mechanism of action. Conclusions Network-based feature selection method and prediction models improve the performance of the drug response prediction. The relations between the genomic features are more robust and stable compared to the correlation between each individual genomic feature and the drug response in high dimension and low sample size genomic datasets.

Sign in / Sign up

Export Citation Format

Share Document