José M. Celaya-Padilla

A comparison of back propagation and Generalized Regression Neural Networks performance in neutron spectrometry.

The process of unfolding the neutron energy spectrum has been subject of research for many years. Monte Carlo, iterative methods, the bayesian theory, the principle of maximum entropy are some of the methods used. The drawbacks associated with traditional unfolding procedures have motivated the research of complementary approaches. Back Propagation Neural Networks (BPNN), have been applied with success in neutron spectrometry and dosimetry domains, however, the structure and learning parameters are factors that highly impact in the networks performance. In ANN domain, Generalized Regression Neural Network (GRNN) is one of the simplest neural networks in term of network architecture and learning algorithm. The learning is instantaneous, requiring no time for training. Opposite to BPNN, a GRNN would be formed instantly with just a 1-pass training on the development data. In the network development phase, the only hurdle is to optimize the hyper-parameter, which is known as sigma, governing the smoothness of the network. The aim of this work was to compare the performance of BPNN and GRNN in the solution of the neutron spectrometry problem. From results obtained it can be observed that despite the very similar results, GRNN performs better than BPNN.

A neutron spectrum unfolding code based on generalized regression artificial neural networks

The most delicate part of neutron spectrometry, is the unfolding process. The derivation of the spectral information is not simple because the unknown is not given directly as a result of the measurements. Novel methods based on Artificial Neural Networks have been widely investigated. In prior works, back propagation neural networks (BPNN) have been used to solve the neutron spectrometry problem, however, some drawbacks still exist using this kind of neural nets, i.e. the optimum selection of the network topology and the long training time. Compared to BPNN, its usually much faster to train a generalized regression neural network (GRNN). Thats mainly because spread constant is the only parameter used in GRNN. Another feature is that the network will converge to a global minimum, provided that the optimal values of spread has been determined and that the dataset adequately represents the problem space. In addition, GRNN are often more accurate than BPNN in the prediction. These characteristics make GRNNs to be of great interest in the neutron spectrometry domain. This work presents a computational tool based on GRNN capable to solve the neutron spectrometry problem. This computational code, automates the pre-processing, training and testing stages using a k-fold cross validation of 3 folds, the statistical analysis and the post-processing of the information, using 7 Bonner spheres rate counts as only entrance data. The code was designed for a Bonner Spheres System based on a 6LiI(Eu) neutron detector and a response matrix expressed in 60 energy bins taken from an International Atomic Energy Agency compilation.

Image Edge Detection Using Fractional Calculus with Feature and Contrast Enhancement

Edge detection is an important aspect of image processing to improve image edge quality. In the literature, there exist various edge detection techniques in spatial and frequency domains that use integer-order differentiation operators. In this paper, we have implemented feature and contrast enhancement of image using Riemann–Liouville fractional differential operator. Based on the direction of strong edge, we have evaluated edge components and carried out a performance analysis based on several well-known metrics. We have also improved the pixel contrast based on foreground and background gray level. Moreover, by theoretical and experimental results, it is observed that the proposed feature and contrast enhancement outperforms the existing methods under comparison. We have discussed that the edge components calculated using fractional derivative can be used for texture and contrast enhancement. This paper is based on fractional-order differentiation operation to detect edges with the help of the directional edge components across eight directions. The experimental comparison results are shown in tabular form and as qualitative texture results. The six experimental input images are used to analyze various performance metrics. The experiments show that for any grayscale image the proposed method outperforms classical edge detection operators.

An Approach to Make Software Testing for Users with Down Syndrome a little more Pleasant

Evaluating software with users implies challenging tasks where users abilities are sometimes taken to the limit. This process might turn a very unpleasant experience to users susceptible to anxiety and stress, such as users with Down syndrome. By consequence the poor performance of unpleasant users generates unreal results. We propose a gamified approach for software testing, which allows maintaining users motivation and engagement with the tests activities, reducing anxiety and stress, having by consequence more reliable results. The gamified approach can be applied on any kind of software testing involving users, since it works over how the activities are presented to participants, once they have been defined according with the test goal. Results of the application, over 10 users with Down syndrome, suggest that the gamified approach maintains the emotional state of users in a positive way even when they made errors, got confused, or were forced to change/stop an activity.

Analyzing Typical Mobile Gestures in mHealth Applications for Users with Down Syndrome

Mobile technology has provided many advantages for all members of the Information Society. Communication, Organization, Transportation, Health, and Entertainment are just a few areas of mobile technology application. Nevertheless, there are still some people who find difficulties using it. Although there are a lot of applications of mHealth available for almost any kind of mobile device, there is still a lack of understanding and attending users’ needs, especially those of users with disabilities. People with Down syndrome have the potential to function as active members of our society, taking care of themselves and their own, having jobs, voting, and so on, but their physical limitations prevent them from handling correctly technological tools that could enhance their performance, including mobile technology. In this paper, we had analyzed how suitable the mHealth applications are for users with Down syndrome. We tested 24 users and analyzed their physical performance in fine-motor movements while developing a set of tasks over a mHealth application. Results showed that the design of a mHealth application for users with Down syndrome must center its interaction with simple gestures as tap and swipe avoiding more complex ones as spread and rotate. This research is a starting point to understand the fundamentals of people with Down syndrome interacting with mobile technology.

Sensitivity and Variability Analysis for Image Denoising Using Maximum Likelihood Estimation of Exponential Distribution

In this paper, we have performed denoising when the pixel values of images are corrupted by Gaussian and Poisson noises. This paper introduces a new class exponential distribution which lies between Poisson and Gamma distributions. The proposed method combines the ion for denoising the pixels and later a minimization using log-likelihood estimation is performed. The characteristic equation is based on various image parameters like mean, variance, mean deviation, distortion index, shape and scale parameters for minimizing the noise and for maximizing image edge strength to enhance overall visual quality of the image. By utilizing the exponential distribution, we can adaptively control the distortion in the image by minimizing Gaussian and Poisson noises in accordance with the image feature. The simulation results indicate that the proposed algorithm is very efficient to strengthen edge information and remove noise. To provide a probabilistic model we have used statistical approximation of mean and variances. Later, we have evaluated sensitivity and variability effect as well on the image restoration. Experiments were conducted on different test images, which were corrupted by different noise levels in order to assess the performance of the proposed algorithm in comparison with standard and other related denoising methods.

A Dynamic Indoor Location Model for Smartphones Based on Magnetic Field: A Preliminary Approach

Due to an increase interest for providing services based on user location, several indoor location approaches based on mobile devices have been proposed recently. This paper focuses on the use of a novel crowdsourcing approach for indoor location of a mobile device that uses social collaboration to improve the accuracy and magnetic field signal as information source using feature extraction and a deterministic method that allows us to include information from new users that improves the fitness of the model. Four phases were included in the methodology: Raw data collection, Data pre-process, Feature extraction and Social collaboration. An experiment was succesfully carried out to test the proposed methodology. On the whole, good results were obtained on computational cost, recalculation time and accuracy improvement.

Automated Image Registration for Knee Pain Prediction in Osteoarthritis: Data from the OAI

Diagnose Knee osteoarthritis (OA) is a very important task, in this work an automated metrics method is used to predict chronic pain. In early stages of OA, changes into joint structures are shown, some of the most common symptoms are; formation of osteophytes, cartilage degradation and joint space reduction, among others. Using public data from the Osteoarthritis initiative (OAI), a set of X-ray images with different Kellgren Lawrence score (K & L) scores were used to determine a relationship between bilateral asymmetry and the radiological evaluation in K & L score with the chronic knee pain. In order to measure the asymmetry between the knees, the right knee was registered to match the left knee, then a series of similarity metrics; mutual information, correlation, and mean square error were computed to correlate the deformation (mismatch) and K & L score with chronic knee pain. Radiological information was evaluated and scored by OAI radiologist groups, all metric of image registration were obtained in an automated way. The results of the study suggest an association between image registration metrics, radiological K & L score with chronic knee pain. Four GLM models wit AUC (> 0.6) and (> 0.7) accuracy random forest classification model was formed with this information to classify the early bony changes with OA chronic knee pain.

Identification and evaluation on diabetic foot injury by computer vision

The diabetic foot is one of the most devastating complications related to diabetic. Its significant transcendence is related to a higher incidence and amputation percentage as well as deaths. Given the fact that laboratory diagnoses trials are both limited and expensive, the most typical alternative is still based on the diseases signs and symptoms. Therefore, the attending physician fills out a questionnaire based on its support instrumental measurements and its own observation (it could be method but not so sure). The aforementioned questionnaire will provide the foundation for the diagnose that also depends on the criteria and the consultants experience. However, for some variables such as the laceration (injury or wound) and-or-location the previous dependency is not acceptable. This paper aims to become the first link to optimize the diabetics foot evaluation through the introduction of digital image processing techniques. Because of the use of advanced object segmentation techniques and a parameter that adjusts the systems sensibility until obtaining the desired results it was possible to apply an algorithm to a series of trial images provided positive results for wound and location detection.