This research study aims to understand the application of Artificial Neural Networks (ANNs) to forecast the Self-Compacting Recycled Coarse Aggregate Concrete (SCRCAC) compressive strength. From different literature, 602 available data sets from SCRCAC mix designs are collected, and the data are rearranged, reconstructed, trained and tested for the ANN model development. The models were established using seven input variables: the mass of cementitious content, water, natural coarse aggregate content, natural fine aggregate content, recycled coarse aggregate content, chemical admixture and mineral admixture used in the SCRCAC mix designs. Two normalization techniques are used for data normalization to visualize the data distribution. For each normalization technique, three transfer functions are used for modelling. In total, six different types of models were run in MATLAB and used to estimate the 28th day SCRCAC compressive strength. Normalization technique 2 performs better than 1 and TANSING is the best transfer function. The best k-fold cross-validation fold is k = 7. The coefficient of determination for predicted and actual compressive strength is 0.78 for training and 0.86 for testing. The impact of the number of neurons and layers on the model was performed. Inputs from standards are used to forecast the 28th day compressive strength. Apart from ANN, Machine Learning (ML) techniques like random forest, extra trees, extreme boosting and light gradient boosting techniques are adopted to predict the 28th day compressive strength of SCRCAC. Compared to ML, ANN prediction shows better results in terms of sensitive analysis. The study also extended to determine 28th day compressive strength from experimental work and compared it with 28th day compressive strength from ANN best model. Standard and ANN mix designs have similar fresh and hardened properties. The average compressive strength from ANN model and experimental results are 39.067 and 38.36 MPa, respectively with correlation coefficient is 1. It appears that ANN can validly predict the compressive strength of concrete.

Reported is the mean of clustering accuracies from 100 runs of Basic NMF together with the standard error of the mean.

All data and figures in the article were simulated by Matlab. The numerical method employed combines the classical spectral method in the temporal domain and the adaptive step size Runge Kutta method in the spatial domain. Herein, we provide all Matlab data and related graph source files for figures 1-6 in the article. This dataset is divided into the following two files. The 'data' file is the Matlab data for Figures 1-6, and the other file is the Matlab source file for Figures 1-6. In the ‘data' file, six files named fig1 (2, 3, 4, 5, 6) were created in the order of the figures, corresponding to the data in figures 1-6. The six files fig1 (2,3,4,5,6) are still named in the other file, according to the order of the figures, corresponding to the source files of figures 1-6. The normalization parameters of Figure 1 are plane wave amplitude A0=1, perturbation frequency Omega=1.5, phase difference(between perturbation signal and pump) phi0=0.5pi, perturbation amplitudes (from top to bottom) delta=0.01, 0.1, and 0.25, respectively. The normalization parameters in Figure 2 are plane wave amplitude A0=1, perturbation frequency Omega=2, phase difference(between perturbation signal and pump) phi0=0.5pi, perturbation amplitudes (from top to bottom) delta=0.01, 0.25, and 0.5, respectively. The normalization parameter in Figure 3 is the plane wave amplitude A0=1, and phase difference(between perturbation signal and pump) of phi0=0.5pi and 0.3pi, respectively. The normalization parameters in Figure 4 are plane wave amplitude A0=1, disturbance frequency Omega=1.5, disturbance amplitude delta=0.1, and the phase difference between the disturbance signal and the pump light (from top to bottom) of phi0=0.1pi, 0.3pi, and 0.5pi, respectively. The normalization parameter in Figure 5 is the plane wave amplitude A0=1, and the perturbation amplitudes are delta=0.1 and 0.01, respectively. The normalization parameters in Figure 6 are plane wave amplitude A0=1, disturbance frequency Omega=2.2, disturbance amplitude delta=0.25, and the phase difference between the disturbance signal and the pump light (from top to bottom) of phi0=0.1pi, 0.3pi, and 0.5pi, respectively.

Companion dataset of the manuscript: Seonyeong Park, Frank J. Brooks, Umberto Villa, Richard Su, Mark A. Anastasio, Alexander A. Oraevsky, "Normalization of optical fluence distribution for three-dimensional functional optoacoustic tomography of the breast," J. Biomed. Opt. 27(3) 036001 (16 March 2022) https://doi.org/10.1117/1.JBO.27.3.036001 This dataset contains an anatomically realistic numerical breast phantom (NBP) to validate optical fluence normalization methods for quantitative and functional optoacoustic tomography (OAT). Specifically, it includes 1) 3D maps of optical absorption coefficient; 2) 3D maps of simulated optical fluence distribution, and 3) 3D OAT reconstructed images at three wavelengths in the near-infrared range. The images were reconstructed from noisy synthetic data using filtered back-projection (FBP). The size of the reconstructed image is 480 × 480 × 240 voxels (120 × 120 × 60 mm3). To simulate the synthetic data, a NBP was created using a computational framework for virtual 3D OAT breast imaging trials, developed by the authors of the reference [Park2020]. Further details of the NBP; functional, optical, and acoustic property assignment; and simulation of optoacoustic signals are in the accompanying paper [Park2022]. The following files are included in this dataset: mu_a_w{757, 800, 850}.mat: Optical absorption coefficient distributions of a NBP at illumination wavelengths of 757 nm, 800 nm, and 850 nm; optical_fluence_w{757, 800, 850}.mat: Optical fluence distributions of a NBP that were simulated at illumination wavelengths of 757 nm, 800 nm, and 850 nm using the MCXLAB software [Fang2009], [Yu2018]; and RECON_NOISY_w{757, 800, 850}_FBP.mat: Images reconstructed from noisy synthetic measurements of a NBP, that were simulated at illumination wavelengths of 757 nm, 800 nm, and 850 nm, using FBP. These data were saved as MATLAB binary files version 5 (extension .mat). They can be imported in MATLAB using the load function. The code of optical fluence normalization is available from the GitHub repository: comp-imaging-sci/optical-fluence-normalization_3d-oat-breast.

Reported is the mean of clustering accuracies from 100 runs of Basic NMF together with the standard error of the mean. Also reported is the p-value produced by a paired two-sided t-test. Note that the proposed method is using ‘max’ normalization and using the filter.

Performance of SVM, NB and PNN for SU-ZS dataset (Reference dataset).

p- value and F- value computation for the data matrices.

F- value of the Datasets with Different Number of Features.

Description of UWB patch antenna used in this research. [27, 28].

Comparison with previous researches (Data from this research is fed into the the existing methods).

Ranking of selected features in Stage 3 based on the F- value.

This research study aims to understand the application of Artificial Neural Networks (ANNs) to forecast the Self-Compacting Recycled Coarse Aggregate Concrete (SCRCAC) compressive strength. From different literature, 602 available data sets from SCRCAC mix designs are collected, and the data are rearranged, reconstructed, trained and tested for the ANN model development. The models were established using seven input variables: the mass of cementitious content, water, natural coarse aggregate content, natural fine aggregate content, recycled coarse aggregate content, chemical admixture and mineral admixture used in the SCRCAC mix designs. Two normalization techniques are used for data normalization to visualize the data distribution. For each normalization technique, three transfer functions are used for modelling. In total, six different types of models were run in MATLAB and used to estimate the 28th day SCRCAC compressive strength. Normalization technique 2 performs better than 1 and TANSING is the best transfer function. The best k-fold cross-validation fold is k = 7. The coefficient of determination for predicted and actual compressive strength is 0.78 for training and 0.86 for testing. The impact of the number of neurons and layers on the model was performed. Inputs from standards are used to forecast the 28th day compressive strength. Apart from ANN, Machine Learning (ML) techniques like random forest, extra trees, extreme boosting and light gradient boosting techniques are adopted to predict the 28th day compressive strength of SCRCAC. Compared to ML, ANN prediction shows better results in terms of sensitive analysis. The study also extended to determine 28th day compressive strength from experimental work and compared it with 28th day compressive strength from ANN best model. Standard and ANN mix designs have similar fresh and hardened properties. The average compressive strength from ANN model and experimental results are 39.067 and 38.36 MPa, respectively with correlation coefficient is 1. It appears that ANN can validly predict the compressive strength of concrete.

Dielectric properties of breast phantom and tumor. [18–20].

This research study aims to understand the application of Artificial Neural Networks (ANNs) to forecast the Self-Compacting Recycled Coarse Aggregate Concrete (SCRCAC) compressive strength. From different literature, 602 available data sets from SCRCAC mix designs are collected, and the data are rearranged, reconstructed, trained and tested for the ANN model development. The models were established using seven input variables: the mass of cementitious content, water, natural coarse aggregate content, natural fine aggregate content, recycled coarse aggregate content, chemical admixture and mineral admixture used in the SCRCAC mix designs. Two normalization techniques are used for data normalization to visualize the data distribution. For each normalization technique, three transfer functions are used for modelling. In total, six different types of models were run in MATLAB and used to estimate the 28th day SCRCAC compressive strength. Normalization technique 2 performs better than 1 and TANSING is the best transfer function. The best k-fold cross-validation fold is k = 7. The coefficient of determination for predicted and actual compressive strength is 0.78 for training and 0.86 for testing. The impact of the number of neurons and layers on the model was performed. Inputs from standards are used to forecast the 28th day compressive strength. Apart from ANN, Machine Learning (ML) techniques like random forest, extra trees, extreme boosting and light gradient boosting techniques are adopted to predict the 28th day compressive strength of SCRCAC. Compared to ML, ANN prediction shows better results in terms of sensitive analysis. The study also extended to determine 28th day compressive strength from experimental work and compared it with 28th day compressive strength from ANN best model. Standard and ANN mix designs have similar fresh and hardened properties. The average compressive strength from ANN model and experimental results are 39.067 and 38.36 MPa, respectively with correlation coefficient is 1. It appears that ANN can validly predict the compressive strength of concrete.

Mean brain to body weight ratio (±SD) and mean percentage of various brain structures in WWCPS, BN and Wistar rats (ie. normalized data).

A post-processed MATLAB structure (MAT-files) that contains stride-normalized data, anthropometric data, and gait events, for able-bodied participants. This data includes kinematic data (i.e., 3D full-body marker data, 3D full-body joint angles, and 3D center of mass), and EMG data (i.e., both not and normalized traces to the maximum value across all strides of a subject of the ERS, RF, VL, BF, ST, TA and GAS for both legs) for all available strides, as well as kinetic data normalized to body mass (i.e., ground reaction forces, joint moments and joint powers) for those strides with clean strikes on the force plate. This data is similar to the xlsx-files.

A post-processed MATLAB structure (MAT-files) that contains stride-normalized data, anthropometric data, and gait events, for able-bodied participants. This data includes kinematic data (i.e., 3D full-body marker data, 3D full-body joint angles, and 3D center of mass), and EMG data (i.e., both not and normalized traces to the maximum value across all strides of a subject of the ERS, RF, VL, BF, ST, TA and GAS for both legs) for all available strides, as well as kinetic data normalized to body mass (i.e., ground reaction forces, joint moments and joint powers) for those strides with clean strikes on the force plate. This data is similar to the xlsx-files.