No abstract

Artificial neural networks have a wide application in many areas of science and engineering and, particularly, in geotechnical problems with some degree of success due to the fact that the mechanical behavior of rocks are not salient. They are highly nonlinear, quite complex and complicated. While applying neural network in such complicated problems, epoch determination is based on hit-and-trail basis mainly. In this paper, the effect of different number of epochs is shown on the network and a method is proposed to determine the optimum number of epoch with the help of self-organized map (SOM) to avoid overtraining of the network. Data distribution is also done with the help of SOM and a statistical analysis is made to show consistency between training and testing dataset for ensuring the optimal model performance.

The prediction of adsorption of cadmium by hematite using an adapted neural fuzzy model and a back propagation artificial neural network was compared. Adsorption was found to depend on the Cd concentration, agitation rate, temperature, pH, and the particle size of the hematite. The adaptive neuro-fuzzy inference system proved to be more efficient in predicting Cd adsorption than a single layered feed forward artificial neural network.

Cerebellum measurements of routinely acquired ultrasound (US) images are commonly used to estimate gestational age and to assess structural abnormalities of the developing central nervous system. Investigating associations between the developing cerebellum and neurodevelopmental outcomes post partum requires standardized cerebellum measurements from large clinical datasets. Such investigations have the potential to identify structural changes that can be used as biomarkers to predict growth and neurodevelopmental outcomes. For this purpose, high throughput, accurate, and unbiased measurements are necessary to replace existing manual, semi-automatic, and automated approaches which are tedious and lack reproducibility and accuracy. In this study, we propose a new deep learning algorithm for automated segmentation of the fetal cerebellum from 2-dimensional (2D) US images. We propose ResU-Net-c a semantic segmentation model optimized for fetal cerebellum structure. We leverage U-Net as a base model with the integration of residual blocks (Res) and introduce dilation convolution in the last two layers to segment the cerebellum (c) from noisy US images. Our experiments used a 5-fold cross-validation with 588 images for training and 146 for testing. Our ResU-Net-c achieved a mean Dice Score Coefficient, Hausdorff Distance, Recall, and Precision of 87.00%, 28.15, 86.00%, and 90.00%, respectively. The superiority of the proposed method over the other U-Net based methods is statistically significant (p < 0.001). Our proposed method can be leveraged to enable high throughput image analysis in clinical research fetal US images and can be employed in the biometric assessment in fetal US images on a larger scale.INDEX TERMS Convolutional neural networks, fetal cerebellum, ResU-Net, segmentation, ultrasound images.

In this study an attempt is made to predict the ratio of muck pile profile before and after the blast, fly rock and total explosive used, based on simple field tests as well blast design parameters. Prediction is done by making three different artificial neural network (ANN) models. Comparative statistical analysis is made among these three networks to ensure their performance suitability. Models of ANN were based on Feed Forward Back Propagation network with training functions – Resilient Backpropagation, One Step Secant and Powell-Beale Restarts. Total numbers of datasets chosen were 92 among which 17 were chosen for testing and validation and the rest were used for the training of networks. Statistical analysis is also made for these datasets. Considering performance for all the outputs, the best results are predicted by Powell-Beale Restarts, with an average percentage error of 5.871% for the ratio of muck pile before and after the blast, 5.335% for fly rocks and 5.775% for total explosive used. These parameters are predicted by number of holes to be blasted, hole diameter, pattern (spacing (m) X burden (m)), total volume of rock in a blast, average depth and total drill depth.