Estimation Of Physiological Parameters Research Articles

Modeling of Shoulder–Elbow Movement with Exponential Parameter Identification During Walking Gaits for Healthy Subjects and Patients with Parkinson’s Disease

Background: This paper aims to complement the latest contribution in the literature that provides estimates of physiological parameters of a dynamic model for the elbow time profile during walking while linking them to a neurodegenerative disorder (Parkinsons’s disease) characterized by motor symptoms. An upper limb model is here proposed in which an active contractile element is included within a model, viewing the arm as a double pendulum system and muscles as represented by a Kelvin–Voight system. All model parameters characterizing both the shoulder and the elbow of each subject are estimated via a gradient-like identifier whose exponential convergence properties are determined by a non-anticipative Lyapunov function, ensuring robustness features. Methods: Joint angle data from different walking subjects (healthy subjects and patients with Parkinson’s disease) have been recorded using an IMU sensor system and compared with the joint angles obtained by means of the proposed model, which was adapted to each subject using available anthropometric knowledge and relying on the estimated parameters. Results: Experiments show that the reconstruction of shoulder and elbow time profiles can be definitely achieved through the proposed procedure with the estimated stiffness parameters turning out to constitute objective and quantitative indices of muscle stiffness (as a pivotal symptom of the pathology), which are able to track changes due to the therapy. Conclusions: The same dynamic model is actually able to capture the main features of the upper limb movement of both (healthy and pathological) walking subjects, with its parameters, in turn, characterizing the nature and progress of the pathology.

A general framework for generative self-supervised learning in non-invasive estimation of physiological parameters using photoplethysmography

Aligning physiological parameter labels with large-scale photoplethysmographic (PPG) data for deep learning is challenging and resource-intensive. While self-supervised representation learning (SSRL) can handle limited annotated data, the challenge lies in learning robust shared representations from vast unlabeled data and integrating various contextual cues to learn distinctive representations. To alleviate these challenges, a generative SSRL framework TS2TC is proposed to collaboratively utilize the temporal, spectrogram, and temporal-spectrogram mixed domains to explore and incorporate the unique features of PPG for universal and non-invasive physiological parameter estimation. Initially, a pretext task named Cross-Temporal Fusion Generative Anchor (CTFGA) is designed, modeling temporal dependencies and reconstructing independent segments at a coarse level to provide robust global feature extraction and local semantic contextual representation. The framework also includes sub-signals from PPG with diverse frequency scales and order derivatives reflecting hemodynamics to facilitate learning shared representations at varying semantic levels. Secondly, an advanced cognitive-inspired dual-process transfer (DPT) strategy is formulated, consisting of prior-dependent autonomous processes and posterior observation reasoning processes, to leverage the independent and integrated advantages of shared and specific representations. Furthermore, TS2TC introduces a novel bilinear temporal-spectrogram fusion method in the mixed domain, aligning latent representations from different domains, and establishing fine-grained contextual interactions at the feature level across multiple sources of information. Extensive experiments on physiological parameter estimation tasks showed that the joint performance of CTFGA and DPT outperforms standard generative learning significantly. TS2TC achieved an average 2.49% improvement in RMSE over the current state-of-the-art estimation methods with only 10% training data.

A model of time-dependent macromolecular and elemental composition of phytoplankton

Phytoplankton Chl:C:N:P ratios are important from both an ecological and a biogeochemical perspective. We show that these elemental ratios can be represented by a phytoplankton physiological model of low complexity that includes major cellular macromolecular pools. In particular, our model resolves time-dependent intracellular pools of chlorophyll, proteins, nucleic acids, carbohydrates/lipids, and N and P storage. Batch culture data for two diatom and two prasinophyte species are used to constrain parameters that represent specific allocation traits and strategies. A key novelty is the simultaneous estimation of physiological parameters for two phytoplankton groups of such different sizes. The number of free parameters is reduced by assuming (i) allometric scaling for maximum uptake rates, (ii) shared half-saturation constants for synthesis of functional macromolecules, (iii) shared exudation rates of functional macromolecules across the species. The rationale behind this assumption is that across the different species, the same or similar processes, enzymes, and metabolites play a role in key physiological processes. For the turnover numbers of macromolecular synthesis and storage exudation rates, differences between diatoms and prasinophytes need to be taken into account to obtain a good fit. Our model fits suggest that the parameters related to storage dynamics dominate the differences in the C:N:P ratios between the different phytoplankton groups. Since descriptions of storage dynamics are still incomplete and imprecise, predictions of C:N:P ratios by phytoplankton models likely have a large uncertainty.

Unified Bayesian network for uncertainty quantification of physiological parameters in dynamic contrast enhanced (DCE) MRI of the liver

Objective. Physiological parameter estimation is affected by intrinsic ambiguity in the data such as noise and model inaccuracies. The aim of this work is to provide a deep learning framework for accurate parameter and uncertainty estimates for DCE-MRI in the liver. Approach. Concentration time curves are simulated to train a Bayesian neural network (BNN). Training of the BNN involves minimization of a loss function that jointly minimizes the aleatoric and epistemic uncertainties. Uncertainty estimation is evaluated for different noise levels and for different out of distribution (OD) cases, i.e. where the data during inference differs strongly to the data during training. The accuracy of parameter estimates are compared to a nonlinear least squares (NLLS) fitting in numerical simulations and in vivo data of a patient suffering from hepatic tumor lesions. Main results. BNN achieved lower root-mean-squared-errors (RMSE) than the NLLS for the simulated data. RMSE of BNN was on overage of all noise levels lower by 33% ± 1.9% for k trans, 22% ± 6% for v e and 89% ± 5% for v p than the NLLS. The aleatoric uncertainties of the parameters increased with increasing noise level, whereas the epistemic uncertainty increased when a BNN was evaluated with OD data. For the in vivo data, more robust parameter estimations were obtained by the BNN than the NLLS fit. In addition, the differences between estimated parameters for healthy and tumor regions-of-interest were significant (p < 0.0001). Significance. The proposed framework allowed for accurate parameter estimates for quantitative DCE-MRI. In addition, the BNN provided uncertainty estimates which highlighted cases of high noise and in which the training data did not match the data during inference. This is important for clinical application because it would indicate cases in which the trained model is inadequate and additional training with an adapted training data set is required.

Estimation of Immunological and Physiological Parameters in Diabetic Patients with COVID-19

Estimation of Immunological and Physiological Parameters in Diabetic Patients with COVID-19

Automated Multi-Wavelength Quality Assessment of Photoplethysmography Signals Using Modulation Spectrum Shape Features.

Photoplethysmography (PPG) is used to measure blood volume changes in the microvascular bed of tissue. Information about these changes along time can be used for estimation of various physiological parameters, such as heart rate variability, arterial stiffness, and blood pressure, to name a few. As a result, PPG has become a popular biological modality and is widely used in wearable health devices. However, accurate measurement of various physiological parameters requires good-quality PPG signals. Therefore, various signal quality indexes (SQIs) for PPG signals have been proposed. These metrics have usually been based on statistical, frequency, and/or template analyses. The modulation spectrogram representation, however, captures the second-order periodicities of a signal and has been shown to provide useful quality cues for electrocardiograms and speech signals. In this work, we propose a new PPG quality metric based on properties of the modulation spectrum. The proposed metric is tested using data collected from subjects while they performed various activity tasks contaminating the PPG signals. Experiments on this multi-wavelength PPG dataset show the combination of proposed and benchmark measures significantly outperforming several benchmark SQIs with improvements of 21.3% BACC (balanced accuracy) for green, 21.6% BACC for red, and 19.0% BACC for infrared wavelengths, respectively, for PPG quality detection tasks. The proposed metrics also generalize for cross-wavelength PPG quality detection tasks.

Tenuazonic acid alters immune and physiological reactions and susceptibility to pathogens in Galleria mellonella larvae.

Tenuazonic acid (TeA) is synthesized by phytopathogenic and opportunistic fungi and is detected in a broad range of foods. This natural compound is of interest in terms of toxicity to animals, but its mechanisms of action on insects are poorly understood. We administered TeA orally at different concentrations (0.2-5.0mg/[gram of a growth medium]) to the model insect Galleria mellonella, with subsequent estimation of physiological, histological, and immunological parameters in different tissues (midgut, fat body, and hemolymph). Susceptibility of the TeA-treated larvae to pathogenic microorganisms Beauveria bassiana and Bacillus thuringiensis was also analyzed. The feeding of TeA to the larvae led to a substation delay of larval growth, apoptosis-like changes in midgut cells, and an increase in midgut bacterial load. A decrease in activities of detoxification enzymes and downregulation of genes Nox, lysozyme, and cecropin in the midgut and/or hemocoel tissues were detected. By contrast, genes gloverin, gallerimycin, and galiomycin and phenoloxidase activity proved to be upregulated in the studied tissues. Hemocyte density did not change under the influence of TeA. TeA administration increased susceptibility of the larvae to B. bassiana but diminished their susceptibility to B. thuringiensis. The results indicate that TeA disturbs wax moth gut physiology and immunity and also exerts a systemic action on this insect. Mechanisms underlying the observed changes in wax moth susceptibility to the pathogens are discussed.

An evaluation of ECG data fusion algorithms for wearable IoT sensors

In wearable sensing, accurate estimation of physiological parameters is paramount, although these signals can be corrupted by noise. The fusion of data from multiple sensor sources has the potential to enhance accuracy, even in the presence of disruptive noise. This paper aims to introduce and compare various existing state-of-the-art and novel data fusion techniques to improve the reliability of heart rate estimation. The comparisons were implemented using the MIT-BIH Arrhythmia database with additive noise signals taken from MIT Noise Stress Test Database. When it comes to the challenging low signal-to-noise ratio (SNR) regions, the Kalman fusion and the α-trim mean filtering approach exhibits the best performance. The Kalman fusion approach dominates when both channels are corrupted, while the α-trim mean filtering elimination algorithm takes the lead when at least one channel is clean. To make the most of these strengths, we have developed an innovative algorithm that can switch between the two fusion methods based on a signal quality indicator (SQI) that serves as a surrogate SNR. This algorithm outperforms the baseline 2-channel RR-interval averaging approach by ≃54% and ≃21% at SNRs of 20dB and −20dB respectively. Moreover, it outperforms other cutting-edge heart rate estimation methods.

Direct machine learning reconstruction of respiratory variation waveforms from resting state fMRI data in a pediatric population

In many functional magnetic resonance imaging (fMRI) studies, respiratory signals are unavailable or do not have acceptable quality due to issues with subject compliance, equipment failure or signal error. In large databases, such as the Human Connectome Projects, over half of the respiratory recordings may be unusable. As a result, the direct removal of low frequency respiratory variations from the blood oxygen level-dependent (BOLD) signal time series is not possible. This study proposes a deep learning-based method for reconstruction of respiratory variation (RV) waveforms directly from BOLD fMRI data in pediatric participants (aged 5 to 21 years old), and does not require any respiratory measurement device. To do this, the Lifespan Human Connectome Project in Development (HCP-D) dataset, which includes respiratory measurements, was used to both train a convolutional neural network (CNN) and evaluate its performance. Results show that a CNN can capture informative features from the BOLD signal time course and reconstruct accurate RV timeseries, especially when the subject has a prominent respiratory event. This work advances the use of direct estimation of physiological parameters from fMRI, which will eventually lead to reduced complexity and decrease the burden on participants because they may not be required to wear a respiratory bellows.

A Review of Wearable Multi-wavelength Photoplethysmography.

Optical pulse detection photoplethysmography (PPG) provides a means of low cost and unobtrusive physiological monitoring that is popular in many wearable devices. However, the accuracy, robustness and generalizability of single-wavelength PPG sensing are sensitive to biological characteristics as well as sensor configuration and placement; this is significant given the increasing adoption of single-wavelength wrist-worn PPG devices in clinical studies and healthcare. Since different wavelengths interact with the skin to varying degrees, researchers have explored the use of multi-wavelength PPG to improve sensing accuracy, robustness and generalizability. This paper contributes a novel and comprehensive state-of-the-art review of wearable multi-wavelength PPG sensing, encompassing motion artifact reduction and estimation of physiological parameters. The paper also encompasses theoretical details about multi-wavelength PPG sensing and the effects of biological characteristics. The review findings highlight the promising developments in motion artifact reduction using multi-wavelength approaches, the effects of skin temperature on PPG sensing, the need for improved diversity in PPG sensing studies and the lack of studies that investigate the combined effects of factors. Recommendations are made for the standardization and completeness of reporting in terms of study design, sensing technology and participant characteristics.

Integrating climate, ecophysiology, and forest cover to estimate the vulnerability of sloths to climate change

AbstractGlobal change imposes multiple challenges on species and, thus, a reliable prediction of current and future vulnerability of species must consider multiple stressors and intrinsic traits of species. Climate, physiology, and forest cover, for example, are required to evaluate threat to thermolabile forest-dependent species, such as sloths (Bradypus spp.; Mammalia: Xenarthra). Here, we estimated future changes in the distribution of three sloth species using a metabolic-hybrid model focused on climate (climatic only, i.e., CO approach) and adding forest cover constraints to distribution of species (climate plus land cover, i.e., CL approach). We used an innovative method to generate estimates of physiological parameters for endotherms, validated with field data. The CF approach predicted a future net expansion of distribution of B. torquatus and B. variegatus, and a future net contraction of distribution of B. tridactylus. The inclusion of forest cover constraints, however, reversed the predictions for B. torquatus, with a predicted net distribution contraction. It also reduced expansion of B. variegatus, although still showing a large net expansion. Thus, B. variegatus is not predicted to be threatened in the future; B. tridactylus emerges as the species most vulnerable to climate change, but with no considerable forest losses, while B. torquatus shows the opposite pattern. Our study highlights the importance of incorporating multiple stressors in predictive models in general. To increase resilience of species to climate change, it is key to control deforestation in the Amazon for B. tridactylus, and to promote reforestation in the Atlantic Forest for B. torquatus.

Impact of Elevated Soil Temperature on Physiological Parameters of Tea Plant Growing Adjacent to Gas Flaring Site

Gas flaring is now recognized as a major environmental problem. Smoking flare may be a significant contributor to overall particulate emissions. It also results in production of excessive heat in the nearby areas. The present investigation aims to study the impact of elevated soil temperature on physiological parameters of tea plant growing adjacent to Gas flaring site. The experimental plot was laid out in 5 x 2 factorial RCBD with two factors viz. different distances from the flare site and two seasons in two different tea gardens adjacent to Kathaloni OCS and Merbil Majuli OCS 6 in Dibrugarh district of Assam in 2019-20. Plant and soil samples were collected and estimation of some select plant physiological parameters was done. Significant increase in soil temperature and decrease in soil moisture content was observed in distance closer to the flare pit. The result of the study revealed that there was a gradual decline in relative turgidity, specific leaf weight, plucking point density, stomatal count, leaf area measurement, polyphenol content, chlorophyll content, caffeine content but an increase in water saturation deficit of the green tea leaves on plants existing at a distance moving closer to the flare pit. A significant increase in rain flushing season as compared to autumn was observed in all the plant parameters except water saturation deficit.

Estimation of some non-enzymatic antioxidants in leaves of Betula platyphylla Sukacz. and Larix sibirica Ledeb.

The study of plant stress tolerance in Mongolia is based on the estimation of morphological and physiological parameters without taking into consideration plant enzyme activity and non-enzymatic antioxidant contents. The latter parameters are more sensitive to various environmental stresses, and their accurate estimation would provide a means to monitor plant health and indirectly determine stress response. This paper presents our study of non-enzymatic antioxidants as chlorophylls (Chl a and Chl b), total carotenoids (TC), ascorbic acid, proline and malondialdehyde, a biomarker of oxidative stress in leaves of two plant species grown in urban and natural environments during the vegetation period. The results indicate that there was a significant decrease in pigments ratio (Chl a : Chl b and TC : (Chl а + Chl в)), and an increase in proline and malondialdehyde contents in urban plants.

Boosted-SpringDTW for Comprehensive Feature Extraction of PPG Signals.

Goal: To achieve high-quality comprehensive feature extraction from physiological signals that enables precise physiological parameter estimation despite evolving waveform morphologies. Methods: We propose Boosted-SpringDTW, a probabilistic framework that leverages dynamic time warping (DTW) and minimal domain-specific heuristics to simultaneously segment physiological signals and identify fiducial points that represent cardiac events. An automated dynamic template adapts to evolving waveform morphologies. We validate Boosted-SpringDTW performance with a benchmark PPG dataset whose morphologies include subject- and respiratory-induced variation. Results: Boosted-SpringDTW achieves precision, recall, and F1-scores over 0.96 for identifying fiducial points and mean absolute error values less than 11.41 milliseconds when estimating IBI. Conclusion: Boosted-SpringDTW improves F1-Scores compared to two baseline feature extraction algorithms by 35% on average for fiducial point identification and mean percent difference by 16% on average for IBI estimation. Significance: Precise hemodynamic parameter estimation with wearable devices enables continuous health monitoring throughout a patients’ daily life.

Applying a Deep Learning Network in Continuous Physiological Parameter Estimation Based on Photoplethysmography Sensor Signals

In this paper, we propose a continuous physiological parameter estimation model based on a deep learning network for photoplethysmography (PPG) sensor signals. Signals of 8-s duration were incorporated into the proposed model in this study for frequent estimation of the systolic blood pressure (BP), diastolic BP, heart rate (HR), and mean arterial pressure of the human body; this facilitated early identification and monitoring of physiological conditions and thus reduced the risk of cardiovascular disease. The proposed model was designed using a convolutional neural network (CNN) and long short-term memory (LSTM) network. This model was trained and validated using the large-scale Multiparameter Intelligent Monitoring in Intensive Care database. The CNN was used to extract features from PPG signals automatically. This automatic extraction replaced the conventional manual feature extraction process. Features with time-series were then analyzed using the LSTM network to estimate physiological parameters. Subsequently, ten-fold cross-validation was conducted to reveal the mean absolute errors ± standard deviations of participants’ systolic BP, diastolic BP, HR, and mean arterial pressure to be 2.54 ± 3.88, 1.59 ± 2.45, 1.62 ± 2.55, and 1.59 ± 2.34 mmHg, respectively. These values meet the standards established by the Association for the Advancement of Medical Instrumentation and the British Hypertension Society. The proposed method facilitates the accurate, continuous monitoring of the BP and HR.

PHYSIOLOGICAL DETERMINANTS OF SHORT TRAIL RUNNING

&lt;p&gt;The recent worldwide popularity of trail running has raised the necessity of studying the physiological profile of this sport. Although trail running races are long distance endurance events, the variety of their terrain, incline and duration prevents the application of the classical predictive model of level running. Thus, the aim of the present study was to investigate the physiological and anthropometric parameters that determine short trail race performance. Twenty-five moderately trained trail runners participated in a 15 km trail running race, consisting of 9 km positive and 6 km negative incline. Four days after the race they followed a laboratory protocol for the measurement and estimation of anthropometric and physiological parameters (maximal oxygen uptake, velocity at maximal oxygen uptake, ventilatory threshold, velocity at ventilatory threshold, running economy, flexibility, muscle power, aerobic capacity). The results revealed high correlations between the 15 km race performance and velocity at maximal oxygen uptake (r = 0.81), ventilatory threshold (r = 0.88), muscle power of knee extensor (r = 0.50 – 0.53), anaerobic capacity (r = 0.65) and body fat percentage (r = 0.7). Another two parameters that were highly correlated with the 15 km mountain trail race performance were both the positive and negative incline time (r = 0.95 and r = 0.96, respectively). Our conclusions confirmed previous findings that performance in trail running cannot be predicted with the same variable model as level running.&lt;/p&gt;&lt;p&gt; &lt;/p&gt;&lt;p&gt;&lt;strong&gt; Article visualizations:&lt;/strong&gt;&lt;/p&gt;&lt;p&gt;&lt;img src="/-counters-/edu_01/0985/a.php" alt="Hit counter" /&gt;&lt;/p&gt;

A geometric misregistration resistant data fusion approach for adding red-edge (RE) and short-wave infrared (SWIR) bands to high spatial resolution imagery

High spatial resolution images have been widely applied to the monitoring of land surfaces during the past decades. However, satellite and unmanned aerial vehicle (UAV) sensors (e.g., IKONOS, and QuickBird) usually only provide visible and near-infrared spectral bands with a high spatial resolution. Such limited spectral bands at a high spatial resolution are not adequate for crucial environmental and agricultural studies that require essential bands (e.g., the red-edge (RE) and the short-wave infrared (SWIR)), such as land cover classification and the estimation of plant physiological parameters. On the contrary, abundant spectral information is commonly available on sensors with a medium spatial resolution (e.g., Landsat, and Sentinel-2). Existing spatial-spectral fusion methods can leverage the unique advantages of these two types of sensors, and produce synthesized images with both spatial and spectral detail. However, most spatial-spectral methods ignore the inherent geometric registration errors when using multi-source data, inevitably introducing the errors into the fusion result. To address the limitation, this study presents a new spatial-spectral fusion method, the Misregistration-Resistant Data Fusion approach (MRDF), to add new spectral bands to the high spatial resolution images from medium resolution images. The proposed method is composed of a local-model-based similar pixel searching and a global partial least square (PLS) regression, and has the following advantages: (1) the local model utilizes neighborhood similar pixels to generate predictions robust against geometric misregistration; (2) the combined model integrates a global PLS regression to improve fusion accuracy for the SWIR bands; (3) the extended-box strategy and spatial filtering can relieve geometric errors when minimizing residuals for the SWIR bands. Satellite and airborne data from three sites were used to compare the performance of MRDF with seven other typical methods. Results showed that MRDF not only produced accurate fusion results with above-average efficiency but also improved the classification accuracy, which substantiated its great potential for various applications.

Beer-Lambert law for optical tissue diagnostics: current state of the art and the main limitations.

.Significance: Beer–Lambert law (BLL) is a widely used tool for contact and remote determination of absorber concentration in various media, including living tissues. Originally proposed in the 18th century as a simple exponential expression, it has survived numerous modifications and updates. The basic assumptions of this law may not be fulfilled in real measurement conditions. This can lead to mistaken or misinterpreted results. In particular, the effects to be additionally taken into account in the tissue measurements include anisotropy, scattering, fluorescence, chemical equilibria, interference, dichroism, spectral bandwidth disagreements, stray radiation, and instrumental effects.Aim: We review the current state of the art and the main limitations of remote tissue diagnostics using the BLL. Historical development of updating this law by taking into account specific additional factors such as light scattering and photon pathlengths in diffuse reflectance is described, along with highlighting the main risks to be considered by interpreting the measured data.Approach: Literature data related to extension and modification of the BLL related to tissue assessment and concentration estimation of specific tissue molecules are collected and analyzed. The main emphasis here is put on the optical measurements of living tissue chromophore concentrations and estimation of physiological parameters, e.g., blood oxygen saturation.Results: Modified expressions of the BLL suitable for several specific cases of living tissue characterization are presented and discussed.Conclusions: Applications of updated/modified Beer–Lambert law (MBLL) with respect to particular measurement conditions are helpful for obtaining more reliable data on the target tissue physiological state and biochemical content. MBLL accounting for the role of scattering in several ways appears to be a successful approach. Extended MBLL and BLL in the time domain form could provide more accurate results, but this requires more time resources to be spent.

Improving the estimation accuracy of SPAD values for maize leaves by removing UAV hyperspectral image backgrounds

ABSTRACT Hyperspectral images collected by unmanned aerial vehicles (UAVs) can provide fine, narrowband spectral information and help realize the accurate estimation of physiological and biochemical crop parameters at the plot scale. However, exposure to different backgrounds will negatively affect crop chlorophyll diagnosis. This research investigated the impact of the background and its degree of influence on the accuracy estimation of soil and plant analyser development (SPAD) values. UAV-based hyperspectral images of 498 maize inbred lines were obtained during the grain filling stage. Twenty vegetation indices (VIs) were calculated. VIs that were sensitive to the SPAD value were screened out by the Boruta algorithm. Estimation accuracies of the SPAD values before and after removing the background were compared and analysed. The results showed that Pearson’s correlation coefficient (r) between VIs with background removal and the SPAD value was higher than that without background removal. The SPAD value estimated with sensitive VIs after removing the background was significantly closer to the measured value than that estimated before removing the background. The verification accuracy of the model with background removal was higher than that without background removal. With background removal, the coefficient of determination (R 2), root mean square error (RMSE) and mean absolute error (MAE) increased by 60.87%, 48.39% and 40.89%, respectively. This study indicates that removal of the background improves the estimation accuracy of the SPAD value for maize leaves. Maize growth parameters could be quickly obtained from UAV hyperspectral images.

Voice Feature Selection to Improve Performance of Machine Learning Models for Voice Production Inversion

Estimation of physiological control parameters of the vocal system from the produced voice outcome has important applications in clinical management of voice disorders . Previously we developed a simulation-based neural network for estimation of vocal fold geometry, mechanical properties, and subglottal pressure from voice outcome features that characterize the acoustics of the produced voice. The goals of this study are to (1) explore the possibility of improving the estimation accuracy of physiological control parameters by including voice outcome features characterizing vocal fold vibration; and (2) identify voice feature sets that optimize both estimation accuracy and robustness to measurement noise. Feedforward neural networks are trained to solve the inversion problem of estimating the physiological control parameters of a three-dimensional body-cover vocal fold model from different sets of voice outcome features that characterize the simulated voice acoustics, glottal flow, and vocal fold vibration. A sensitivity analysis is then performed to evaluate the contribution of individual voice features to the overall performance of the neural networks in estimating the physiologic control parameters. While including voice outcome features characterizing vocal fold vibration increases estimation accuracy, it also reduces the network's robustness to measurement noise, due to high sensitivity of network performance to voice outcome features measuring the absolute amplitudes of the glottal flow and area waveforms, which are also difficult to measure accurately in practical applications. By excluding such glottal flow-based features and replacing glottal area-based features by their normalized counterparts, we are able to significantly improve both estimation accuracy and robustness to noise. We further show that similar estimation accuracy and robustness can be achieved with an even smaller set of voice outcome features by excluding features of small sensitivity.