Estimating biological age by hematological blood parameters

Abstract. For the estimation of the biological age (BA) of people based on hematological parameters of the clinical blood test there were used MLR and Deep Neural Networks. In the archive of the Institute of Gerontology NAMS of Ukraine there were selected people aged from 20 up to 90 years (440 men and 504 women), who had all hematological parameters within normal limits. When using the MLR method, the multiple correlation coefficients (R) have low values for both men (0.37) and women (0.38). The use of Deep Neural Networks has given good results. The values of the correlation coefficients between BA and chronological age were 0.92 for men and 0.79 for women. The average absolute error in determining BA was 3.68 years for the men and 6.55 years for the women. The developed method for assessing hematological age can be used in clinical practice to identify people with the risk of developing hematological pathology, as well as in population researches. Keywords: biological age, hematological blood parameters, deep neural network

Download Full-text

Automated Analysis of Medial Gastrocnemius Muscle-Tendon Junction Displacements During Isolated Contractions and Walking Using Deep Neural Networks

10.1101/2020.09.29.317529 ◽

2020 ◽

Author(s):

Rebecca L. Krupenevich ◽

Callum J. Funk ◽

Jason R. Franz

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

Automated Analysis ◽

Absolute Error ◽

Clinical Diagnostics ◽

Treadmill Walking ◽

Medial Gastrocnemius ◽

Ultrasound Images ◽

Time Saving ◽

Software Packages

AbstractDirect measurement of muscle-tendon junction (MTJ) position is important for understanding dynamic tendon behavior and muscle-tendon interaction in healthy and pathological populations. Traditionally, obtaining MTJ position during functional activities is accomplished by manually tracking the position of the MTJ in cine B-mode ultrasound images – a laborious and time-consuming process. Recent advances in deep learning have facilitated the availability of user-friendly open-source software packages for automated tracking. However, these software packages were originally intended for animal pose estimation and have not been widely tested on ultrasound images. Therefore, the purpose of this paper was to evaluate the efficacy of deep neural networks to accurately track medial gastrocnemius MTJ positions in cine B-mode ultrasound images across tasks spanning controlled loading during isolated contractions to physiological loading during treadmill walking. Cine B-mode ultrasound images of the medial gastrocnemius MTJ were collected from 15 subjects (6M/9F, 23 yr, 71.9 kg, 1.8 m) during treadmill walking at 1.25 m/s and during maximal voluntary isometric plantarflexor contractions (MVICs). Five deep neural networks were trained using 480 labeled images collected during walking, and were then used to predict MTJ position in images from novel subjects 1) during walking (novel-subject), and 2) during MVICs (novel-condition). We found an average mean absolute error of 1.26±1.30 mm and 2.61±3.31 mm in the novel-subject and novel-condition evaluations, respectively. We believe this approach to MTJ position tracking is an accessible and time-saving solution, with broad applications for many fields, such as rehabilitation or clinical diagnostics.

Download Full-text

Towards a Whole Body [18F] FDG Positron Emission Tomography Attenuation Correction Map Synthesizing using Deep Neural Networks

Journal of Computer Science and Technology ◽

10.24215/16666038.21.e4 ◽

2021 ◽

Vol 21 (1) ◽

pp. e4

Author(s):

Ramiro Germán Rodríguez Colmeiro ◽

Claudio Verrastro ◽

Daniel Minsky ◽

Thomas Grosges

Keyword(s):

Neural Networks ◽

Positron Emission Tomography ◽

Attenuation Correction ◽

Deep Neural Networks ◽

Absolute Error ◽

Emission Tomography ◽

Whole Body ◽

Multiple Sources ◽

Attenuation Map ◽

Positron Emission

The correction of attenuation effects in Positron Emission Tomography (PET) imaging is fundamental to obtain a correct radiotracer distribution. However direct measurement of this attenuation map is not error-free and normally results in additional ionization radiation dose to the patient. Here, we explore the task of whole body attenuation map generation using 3D deep neural networks. We analyze the advantages thar an adversarial network training cand provide to such models. The networks are trained to learn the mapping from non attenuation corrected [18 ^F]-fluorodeoxyglucose PET images to a synthetic Computerized Tomography (sCT) and also to label the input voxel tissue. Then the sCT image is further refined using an adversarial training scheme to recover higher frequency details and lost structures using context information. This work is trained and tested on public available datasets, containing several PET images from different scanners with different radiotracer administration and reconstruction modalities. The network is trained with 108 samples and validated on 10 samples. The sCT generation was tested on 133 samples from 8 distinct datasets. The resulting mean absolute error of the networks is 90±20 and 103±18HU and a peak signal to noise ratio of 19.3±1.7 dB and 18.6±1.5, for the base model and the adversarial model respectively. The attenuation correction is tested by means of attenuation sinograms, obtaining a line of response attenuation mean error lower than 1% with a standard deviation lower than 8%. The proposeddeep learning topologies are capable of generating whole body attenuation maps from uncorrected PET image data. Moreover, the accuracy of both methods holds in the presence of data from multiple sources and modalities and are trained on publicly available datasets. Finally, while the adversarial layer enhances visual appearance of the produced samples, the 3D U-Net achieves higher metric performance

Download Full-text

Deep Neural Network for Visual Stimulus-Based Reaction Time Estimation Using the Periodogram of Single-Trial EEG

Sensors ◽

10.3390/s20216090 ◽

2020 ◽

Vol 20 (21) ◽

pp. 6090

Author(s):

Mohammad Samin Nur Chowdhury ◽

Arindam Dutta ◽

Matthew Kyle Robison ◽

Chris Blais ◽

Gene Arnold Brewer ◽

...

Keyword(s):

Neural Network ◽

Neural Networks ◽

Reaction Time ◽

Visual Stimulus ◽

High Performance ◽

Deep Neural Networks ◽

Robust Regression ◽

Correlation Coefficients ◽

Time Estimation ◽

Single Trial

Multiplexed deep neural networks (DNN) have engendered high-performance predictive models gaining popularity for decoding brain waves, extensively collected in the form of electroencephalogram (EEG) signals. In this paper, to the best of our knowledge, we introduce a first-ever DNN-based generalized approach to estimate reaction time (RT) using the periodogram representation of single-trial EEG in a visual stimulus-response experiment with 48 participants. We have designed a Fully Connected Neural Network (FCNN) and a Convolutional Neural Network (CNN) to predict and classify RTs for each trial. Though deep neural networks are widely known for classification applications, cascading FCNN/CNN with the Random Forest model, we designed a robust regression-based estimator to predict RT. With the FCNN model, the accuracies obtained for binary and 3-class classification were 93% and 76%, respectively, which further improved with the use of CNN (94% and 78%, respectively). The regression-based approach predicted RTs with correlation coefficients (CC) of 0.78 and 0.80 for FCNN and CNN, respectively. Investigating further, we found that the left central as well as parietal and occipital lobes were crucial for predicting RT, with significant activities in the theta and alpha frequency bands.

Download Full-text

Friction Factor Prediction for Newtonian and Non-Newtonian Fluids in Pipe Flows Using Neural Networks

International Journal of Food Engineering ◽

10.2202/1556-3758.1079 ◽

2007 ◽

Vol 3 (1) ◽

Cited By ~ 7

Author(s):

Gauri S. Mittal ◽

Jixian Zhang

Keyword(s):

Neural Networks ◽

Yield Stress ◽

Power Law ◽

Absolute Error ◽

Newtonian Fluids ◽

Fluid Viscosity ◽

Prediction Errors ◽

Average Absolute Error ◽

Bingham Plastic ◽

Power Law Fluids

The friction factors (f) for Newtonian, power law, Bingham plastic and Herschel-Bulkely fluids were predicted after developing and training four neural networks (NN). Three and four layer NN and Wardnet slab were used for f predictions. When average velocity (u), pipe diameter (D), fluid density and fluid viscosity were used for predicting f values for Newtonian fluids, average absolute error was only 0.00004 with standard deviation of 0.00050 and correlation coefficient (r) of 0.9981. When using flow behaviour index (n),u, D, density and consistency coefficient (k) as inputs of an NN for power law fluids, the average absolute error of predicting f was 0.0116 with r of 0.9998. For prediction of f using yield stress, u, D, density and plastic viscosity as inputs to an NN for Bingham plastic fluids, the average absolute error was 0.0044 with r of 0.9961. The average absolute error was 0.0169 with r of 0.9996 for the prediction of f taking n, yield stress, u, D, density and k as inputs to an NN for Herschel-Bulkely fluids. Inputs except n and density and output were transformed on a logarithmic base to 10 scale. Prediction using log f or extension of f limit reduced prediction errors.

Download Full-text

Water temperature fluctuations as a key driver of cetacean pox (tattoo) lesions in bottlenose dolphins Tursiops truncatus

Diseases of Aquatic Organisms ◽

10.3354/dao03472 ◽

2020 ◽

Vol 139 ◽

pp. 69-79

Author(s):

LA Croft ◽

R Laughlin ◽

M Manley ◽

HH Nollens

Keyword(s):

Water Temperature ◽

Food Consumption ◽

Temperature Increase ◽

Complete Blood Count ◽

Hematological Parameters ◽

Bottlenose Dolphins ◽

Skin Lesions ◽

Blood Parameters ◽

Normal Limits ◽

Key Driver

Dolphin tattoo lesions are superficial non-raised skin lesions caused by poxviruses. Their presentation can vary but typical lesions in bottlenose dolphins are circular to ovoid with concentric rings of black stippling. These lesions have at times been suggested as an indicator of overall dolphin health and welfare. This study explored the effect of water temperature on the extent of tattoo lesions in 25 dolphins, along with established hematological health parameters and food consumption. Study animals consisted of 9 males and 16 females with dolphin tattoo lesions ranging in age from 2 to 45 yr. A significant decrease (p < 0.01) in extent and appearance of tattoo lesions was documented following increase in water temperature from 21-24°C (70-75°F) to 25.5-26.5°C (78-80°F). Reduction in tattoo lesions could be noted as early as 2-3 wk following water temperature increase. Marked reduction to complete resolution of tattoo lesions was reproducibly seen 5-6 wk post temperature increase. Food consumption following temperature increase was variable: decrease in intake was noted in 48% of dolphins, increase in intake in 52%. Routine blood parameters (complete blood count, serum chemistry panel, fibrinogen, erythrocyte sedimentation rate) remained within normal limits. A significant increase (p < 0.05) in extent and appearance of tattoo lesions was documented in 3 dolphins as early as 4-5 wk following a decrease in water temperature from 26.1°C (79°F) to 21°C (70°F). Water temperature is a key environmental parameter affecting cetacean pox (‘tattoo’) lesions in bottlenose dolphins. The absence of changes in hematological parameters along with lack of correlation between extent of pox lesions and food intake indicates that dolphin tattoo lesions are not an appropriate indicator of overall health.

Download Full-text

Review on Deep Neural Networks Applied to Low-Frequency NILM

Energies ◽

10.3390/en14092390 ◽

2021 ◽

Vol 14 (9) ◽

pp. 2390

Author(s):

Patrick Huber ◽

Alberto Calatroni ◽

Andreas Rumsch ◽

Andrew Paice

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

Low Frequency ◽

Absolute Error ◽

Performance Comparison ◽

Large Field ◽

Data Sampling ◽

Research Gaps ◽

Generative Adversarial Network ◽

Task Learning

This paper reviews non-intrusive load monitoring (NILM) approaches that employ deep neural networks to disaggregate appliances from low frequency data, i.e., data with sampling rates lower than the AC base frequency. The overall purpose of this review is, firstly, to gain an overview on the state of the research up to November 2020, and secondly, to identify worthwhile open research topics. Accordingly, we first review the many degrees of freedom of these approaches, what has already been done in the literature, and compile the main characteristics of the reviewed publications in an extensive overview table. The second part of the paper discusses selected aspects of the literature and corresponding research gaps. In particular, we do a performance comparison with respect to reported mean absolute error (MAE) and F1-scores and observe different recurring elements in the best performing approaches, namely data sampling intervals below 10 s, a large field of view, the usage of generative adversarial network (GAN) losses, multi-task learning, and post-processing. Subsequently, multiple input features, multi-task learning, and related research gaps are discussed, the need for comparative studies is highlighted, and finally, missing elements for a successful deployment of NILM approaches based on deep neural networks are pointed out. We conclude the review with an outlook on possible future scenarios.

Download Full-text

Modeling Deep Neural Networks to Learn Maintenance and Repair Costs of Educational Facilities

Buildings ◽

10.3390/buildings11040165 ◽

2021 ◽

Vol 11 (4) ◽

pp. 165

Author(s):

Jimyong Kim ◽

Sangguk Yum ◽

Seunghyun Son ◽

Kiyoung Son ◽

Junseo Bae

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

Holistic Approach ◽

Absolute Error ◽

Facilities Management ◽

Educational Facilities ◽

Modeling Framework ◽

Maintenance And Repair ◽

Repair Costs ◽

Data Points

Educational facilities hold a higher degree of uncertainty in predicting maintenance and repair costs than other types of facilities. Moreover, achieving accurate and reliable maintenance and repair costs is essential, yet very little is known about a holistic approach to learning them by incorporating multi-contextual factors that affect maintenance and repair costs. This study fills this knowledge gap by modeling and validating deep neural networks to efficiently and accurately learn maintenance and repair costs, drawing on 1213 high-confidence data points. The developed model learns and generalizes claim payout records on the maintenance and repair costs from sets of facility asset information, geographic profiles, natural hazard records, and other causes of financial losses. The robustness of the developed model was tested and validated by measuring the root mean square error and mean absolute error values. This study attempted to propose an analytical modeling framework that can accurately learn various factors, significantly affecting the maintenance and repair costs of educational facilities. The proposed approach can contribute to the existing body of knowledge, serving as a reference for the facilities management of other functional types of facilities.

Download Full-text

Spatially Disaggregated Car Ownership Prediction Using Deep Neural Networks

Future Transportation ◽

10.3390/futuretransp1010008 ◽

2021 ◽

Vol 1 (1) ◽

pp. 113-133

Author(s):

James Dixon ◽

Sofia Koukoura ◽

Christian Brand ◽

Malcolm Morgan ◽

Keith Bell

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

Absolute Error ◽

Stochastic Gradient Descent ◽

Car Ownership ◽

Support Vector ◽

Geographic Datasets ◽

Dissipation Model ◽

The Uk ◽

Over Time

Predicting car ownership patterns at high spatial resolution is key to understanding pathways for decarbonisation—via electrification and demand reduction—of the private vehicle fleet. As the factors widely understood to influence car ownership are highly interdependent, linearised regression models, which dominate previous work on spatially explicit car ownership modelling in the UK, have shortcomings in accurately predicting the relationship. This paper presents predictions of spatially disaggregated car ownership—and change in car ownership over time—in Great Britain (GB) using deep neural networks (NNs) with hyperparameter tuning. The inputs to the models are demographic, socio-economic and geographic datasets compiled at the level of Census Lower Super Output Areas (LSOAs)—areas covering between 300 and 600 households. It was found that when optimal hyperparameters are selected, these neural networks can predict car ownership with a mean absolute error of up to 29% lower than when formulating the same problem as a linear regression; the results from NN regression are also shown to outperform three other artificial intelligence (AI)-based methods: random forest, stochastic gradient descent and support vector regression. The methods presented in this paper could enhance the capability of transport/energy modelling frameworks in predicting the spatial distribution of vehicle fleets, particularly as demographics, socio-economics and the built environment—such as public transport availability and the provision of local amenities—evolve over time. A particularly relevant contribution of this method is that by coupling it with a technology dissipation model, it could be used to explore the possible effects of changing policy, behaviour and socio-economics on uptake pathways for electric vehicles —cited as a vital technology for meeting Net Zero greenhouse gas emissions by 2050.

Download Full-text