Mean Absolute Error Measure Research Articles

Automatic detection of posterior superior alveolar artery in dental cone-beam CT images using a deeply supervised multi-scale 3D network.

This study aimed to develop a robust and accurate deep learning network for detecting the posterior superior alveolar artery (PSAA) in dental cone-beam CT (CBCT) images, focusing on the precise localization of the centre pixel as a critical centreline pixel. PSAA locations were manually labelled on dental CBCT data from 150 subjects. The left maxillary sinus images were horizontally flipped. In total, 300 datasets were created. Six different deep learning networks were trained, including 3D U-Net, deeply supervised 3D U-Net (3D U-Net DS), multi-scale deeply supervised 3D U-Net (3D U-Net MSDS), 3D Attention U-Net, 3D V-Net, and 3D Dense U-Net. The performance evaluation involved predicting the centre pixel of the PSAA. This was assessed using mean absolute error (MAE), mean radial error (MRE), and successful detection rate (SDR). The 3D U-Net MSDS achieved the best prediction performance among the tested networks, with an MAE measurement of 0.696 ± 1.552 mm and MRE of 1.101 ± 2.270 mm. In comparison, the 3D U-Net showed the lowest performance. The 3D U-Net MSDS demonstrated a SDR of 95% within a 2 mm MAE. This was a significantly higher result than other networks that achieved a detection rate of over 80%. This study presents a robust deep learning network for accurate PSAA detection in dental CBCT images, emphasizing precise centre pixel localization. The method achieves high accuracy in locating small vessels, such as the PSAA, and has the potential to enhance detection accuracy and efficiency, thus impacting oral and maxillofacial surgery planning and decision-making.

Decomposed Dissimilarity Measure for Evaluation of Digital Image Denoising

A new approach to the evaluation of digital image denoising algorithms is presented. In the proposed method, the mean absolute error (MAE) is decomposed into three components that reflect the different cases of denoising imperfections. Moreover, aim plots are described, which are designed to be a very clear and intuitive form of presentation of the new decomposed measure. Finally, examples of the application of the decomposed MAE and the aim plots in the evaluation of impulsive noise removal algorithms are presented. The decomposed MAE measure is a hybrid of the image dissimilarity measure and detection performance measures. It provides information about sources of errors such as pixel estimation errors, unnecessary altered pixels, or undetected and uncorrected distorted pixels. It measures the impact of these factors on the overall correction performance. The decomposed MAE is suitable for the evaluation of algorithms that perform a detection of the distortion that affects only a certain fraction of the image pixels.

Implementation of 1D convolutional neural network for improvement remote photoplethysmography measurement

<span lang="EN-US">Remote photoplethysmography (rPPG) for non-contact heart rate measurement has been widely developed and shows good development. However, motion artifact due to changes in illumination and subject movement is still the main problem. Especially when measurements are taken in real conditions. In these conditions, it will be vulnerable to rPPG signal readings with poor signal quality. So, in this paper, it is proposed to classify the signal quality using one dimensional convolutional neural network (1D CNN). The classification is carried out based on the extraction of the temporal features of the rPPG signal that has been obtained from the plane orthogonal to skin algorithm and the magnitude of the subject's movement when measured. The classification results are entered into a compensated network if the signal obtained shows moderate quality. The compensated network will provide a more accurate estimate of hr value. The test was carried out using a dataset of 10 subjects, each measured with 3 different types of illumination. In the experiments conducted, the system's performance showed an improvement compared to the POS algorithm alone. The experiment found that the mean absolute error measurement was 2.78, and the mean error was relative at 3.67%.</span>

Salient Object Detection by LTP Texture Characterization on Opposing Color Pairs under SLICO Superpixel Constraint.

The effortless detection of salient objects by humans has been the subject of research in several fields, including computer vision, as it has many applications. However, salient object detection remains a challenge for many computer models dealing with color and textured images. Most of them process color and texture separately and therefore implicitly consider them as independent features which is not the case in reality. Herein, we propose a novel and efficient strategy, through a simple model, almost without internal parameters, which generates a robust saliency map for a natural image. This strategy consists of integrating color information into local textural patterns to characterize a color micro-texture. It is the simple, yet powerful LTP (Local Ternary Patterns) texture descriptor applied to opposing color pairs of a color space that allows us to achieve this end. Each color micro-texture is represented by a vector whose components are from a superpixel obtained by the SLICO (Simple Linear Iterative Clustering with zero parameter) algorithm, which is simple, fast and exhibits state-of-the-art boundary adherence. The degree of dissimilarity between each pair of color micro-textures is computed by the FastMap method, a fast version of MDS (Multi-dimensional Scaling) that considers the color micro-textures’ non-linearity while preserving their distances. These degrees of dissimilarity give us an intermediate saliency map for each RGB (Red–Green–Blue), HSL (Hue–Saturation–Luminance), LUV (L for luminance, U and V represent chromaticity values) and CMY (Cyan–Magenta–Yellow) color space. The final saliency map is their combination to take advantage of the strength of each of them. The MAE (Mean Absolute Error), MSE (Mean Squared Error) and F measures of our saliency maps, on the five most used datasets show that our model outperformed several state-of-the-art models. Being simple and efficient, our model could be combined with classic models using color contrast for a better performance.

Modelling of biodiesel blend using optimised deep belief network: blending waste cooking oil methyl ester with tyre pyrolysis oil

This study introduces a new biodiesel blend as an alternative for diesel using waste cooking oil methyl ester by adding tyre pyrolysis oil and cerium oxide. Despite the conventional biodiesel blending models, this study made an effort to efficiently measure the prediction rate of these blended fuels by modelling through the deep belief network (DBN). To attain the accurate prediction, this study moves on with the new logic of optimal tuning of the count of hidden neurons in DBN. The optimal selection is carried out by introducing a new algorithm named lioness updated crow search algorithm (LCSA), which hybrids the concept of the lion algorithm (LA) and crow search algorithm (CSA). Finally, the proposed work is analysed and compared over other conventional models with respect to emission analysis and error analysis. From the analysis, the proposed model in terms of mean deviation (MD) measure has gained betterment and is 75.57, 17.71, 85.55, and 74.19% better than grey wolf optimiser (GWO), whale optimisation algorithm (WOA), LA, and CSA, respectively. For the mean absolute error measure, the implemented model is 42.38, 24.42, 43.53 and 36.72% improved than GWO, WOA, LA, and CSA, respectively.

Landslide susceptibility modelling using integrated evidential belief function based logistic regression method: A study from Southern Western Ghats, India

The present study aims to develop a spatially integrated evidential belief function based logistic regression model (EBF-LR) for landslide susceptibility mapping in a naturally sloping terrain of Southern Western Ghats in Kerala, India. For this, a landslide inventory map of 83 previous landslides was prepared using satellite imageries and verified in field check. Thereafter the landslide inventory was randomly divided into 70%–30% basis for model training and testing. Twelve landslide conditioning factors viz., lithology, land use/land cover, NDVI, slope angle, slope aspect, profile curvature, distance to stream, distance to roads, distance to lineaments, soil texture, topographic wetness index and average annual rainfall were considered for landslide susceptibility modelling. The resultant susceptibility maps were validated using receiver operating characteristics curve (ROC) with area under the curve (AUC) value, sensitivity, specificity, kappa index, mean absolute error (MAE) and root mean square error (RMSE). Analysis shows that the integrated EBF-LR model outweigh other conventional bivariate and multivariate approaches with a ROC-AUC value of 0.935, Kappa score 0.719, sensitivity 0.885, specificity 0.833 and least RMSE of 0.456 in the validation stage. The study also reveals that anthropogenic disturbances have a significant role on landslide initiation in the study area. Considering the ROC- AUC, MAE, RMSE and other validation measures, the accuracy of the proposed landslide susceptibility model is satisfactory and can be used for future land use planning and landslide mitigation in the study area.

PCA Forecast Averaging—Predicting Day-Ahead and Intraday Electricity Prices

Recently, the development in combining point forecasts of electricity prices obtained with different length of calibration windows have provided an extremely efficient and simple tool for improving predictive accuracy. However, the proposed methods are strongly dependent on expert knowledge and may not be directly transferred from one to another model or market. Hence, we consider a novel extension and propose to use principal component analysis (PCA) to automate the procedure of averaging over a rich pool of predictions. We apply PCA to a panel of over 650 point forecasts obtained for different calibration windows length. The robustness of the approach is evaluated with three different forecasting tasks, i.e., forecasting day-ahead prices, forecasting intraday ID3 prices one day in advance, and finally very short term forecasting of ID3 prices (i.e., six hours before delivery). The empirical results are compared using the Mean Absolute Error measure and Giacomini and White test for conditional predictive ability (CPA). The results indicate that PCA averaging not only yields significantly more accurate forecasts than individual predictions but also outperforms other forecast averaging schemes.

Spectral Subtraction Denoising Preprocessing Block to Improve Slow Cortical Potential Based Brain–Computer Interface

The measured electroencephalographic (EEG) signals used in brain–computer interface (BCI) are usually contaminated with several additive random and physiological noise components. Even though several preprocessing strategies were proposed in the literature to extract useful EEG signal components for further processing and analysis, their performance is yet to meet practical needs to boost BCI applications performance. Of particular interest among those methods are the ones addressing random noise suppression that could help enhance the performance of low-cost BCI systems/headsets. A preprocessing block for signal denoising of slow cortical potential (SCP) that provides efficient noise removal and better classification accuracy is presented. This method is a variant of the spectral subtraction signal denoising whereby the noise power spectrum is estimated and removed adaptively. This method is applied to each EEG channel separately thus providing a preprocessing block that could be integrated with the existing spatial/temporal preprocessing methods. The impact of this method on classification accuracy is studied and compared to the conventional wavelet shrinkage method. The proposed method is verified using experimental data from the BCI competition II data set whereby a performance boost was obtained with the new preprocessing block. The improvement was quantitatively assessed using mean square error and mean absolute error measures and was also shown to be statistically significant. Moreover, spectral subtraction denoising is shown to have less computational complexity than wavelet shrinkage based methods. The new preprocessing block for SCP-based BCI signals based on spectral subtraction provides significant improvement in performance when added and offers an adaptive yet less computationally expensive alternative to existing methods such as wavelet shrinkage based denoising method. Also, given its independent channel processing, it shows potential for seamless integration into conventional processing chain for different BCI applications.

Macroeconomic Determinants of Volatility for the Gold Price in Ethiopia: The Application of GARCH and EWMA Volatility Models

Modelling and forecasting of commodity price volatility has important applications for asset management, portfolio analysis and risk assessment due to the simple fact that volatility has informational content and contains signals of the market information flow. This article models and forecasts the gold price volatility using the exponentially weighted moving average (EWMA) and the generalized autoregressive conditional heteroscedasticity (GARCH) models for the period from 1998 to 2014. The gold series shows the classical characteristics of financial time series, such as leptokurtic distributions, data dependence and strong serial correlation in squared returns. Hence, the series can be modelled using both EWMA and GARCH-type models. Among the GARCH-type models, GARCH-M(2,2) with Student’s t distribution for the residuals was found to be the best-fit model. Moreover, the manuscript finds that interest rates, exchange rates and crude oil prices have a significant impact on gold volatility. The risk premium effect is found to be positive and statistically significant, suggesting increased volatility is followed by a higher mean. Finally, a comparison is made between the GARCH and the EWMA models. Using the relative mean squared error and mean absolute error measures, the empirical result suggests that GARCH models with explanatory variables are superior for volatility forecasting.

Dead Sea Starvation: Towards Enhanced Monitoring of Water Resources by Modeling Meteorological Variables and Remote Sensing Data

Meteorological metrics have been used for weather forecasting and climate prediction. Remote sensing images proved to be a valuable resource to represent the terrain of earth’s surface. Recently, there has been extensive research to model changes on the earth’s landscape including water bodies using remote sensing images. Meanwhile, meteorological data have been used mainly to model climate changes. This research tries to leverage both resources to achieve enhanced monitoring of the Dead Sea shrinkage: first, an attempt to model the relation between several meteorological variables and Dead Sea shrinkage using machine learning; second, formulating Dead Sea shrinkage in terms of water level and surface area using data extraction from remote sensing images; finally, confronting the two models to derive a novel approach for predicting Dead Sea shrinkage based on spatiotemporal images and meteorological measures. The main machine learning algorithms for modeling the water shrinkage in this empirical research are Decision Table, Linear Regression, and Multi Layer Perceptron Neural Networks. The Mean Absolute Error measure of the best model is 1.743 and 0.015. It is challenging to model the relation between meteorological variables and the water level. However, the obtained results are promising to formulate a model of the water level decline rate, which in its turn will be an essential tool for estimating the consumption limits and inflow needs to save the Dead Sea.

A hybrid approach for index tracking with practical constraints

Index tracking is a popular way for passive fund management, which aims to reproduce the performance of a market index by investing in a subset of the constituents of the index. The formulation of index tracking with some realistic constraints always leads to an optimization problem that is very hard to solve. In this paper, we propose an approximate formulation to the original optimization problem and analyze the approximation error bound. It is shown that the approximation can be reasonably close to the original problem. We consider both cases where the mean absolute error and mean square error are used as the tracking error measurements. The mean absolute error measurement results in a mixed-integer linear programming problem and can be solved by some standard solvers, such as CPLEX. The mean square error measurement leads to a mixed-integer quadratic programming problem. An efficient hybrid heuristic method is given to solve this problem. We do some numerical experiments by the use of five data sets from OR-Library. The results are promising.