RESUMEN

Introduction: Wearable assistive devices for the visually impaired whose technology is based on video camera devices represent a challenge in rapid evolution, where one of the main problems is to find computer vision algorithms that can be implemented in low-cost embedded devices. Objectives and Methods: This work presents a Tiny You Only Look Once architecture for pedestrian detection, which can be implemented in low-cost wearable devices as an alternative for the development of assistive technologies for the visually impaired. Results: The recall results of the proposed refined model represent an improvement of 71% working with four anchor boxes and 66% with six anchor boxes compared to the original model. The accuracy achieved on the same data set shows an increase of 14% and 25%, respectively. The F1 calculation shows a refinement of 57% and 55%. The average accuracy of the models achieved an improvement of 87% and 99%. The number of correctly detected objects was 3098 and 2892 for four and six anchor boxes, respectively, whose performance is better by 77% and 65% compared to the original, which correctly detected 1743 objects. Discussion: Finally, the model was optimized for the Jetson Nano embedded system, a case study for low-power embedded devices, and in a desktop computer. In both cases, the graphics processing unit (GPU) and central processing unit were tested, and a documented comparison of solutions aimed at serving visually impaired people was performed. Conclusion: We performed the desktop tests with a RTX 2070S graphics card, and the image processing took about 2.8 ms. The Jetson Nano board could process an image in about 110 ms, offering the opportunity to generate alert notification procedures in support of visually impaired mobility.

RESUMEN

The pre-diagnosis of cancer has been approached from various perspectives, so it is imperative to continue improving classification algorithms to achieve early diagnosis of the disease and improve patient survival. In the medical field, there are data that, for various reasons, are lost. There are also datasets that mix numerical and categorical values. Very few algorithms classify datasets with such characteristics. Therefore, this study proposes the modification of an existing algorithm for the classification of cancer. The said algorithm showed excellent results compared with classical classification algorithms. The AISAC-MMD (Mixed and Missing Data) is based on the AISAC and was modified to work with datasets with missing and mixed values. It showed significantly better performance than bio-inspired or classical classification algorithms. Statistical analysis established that the AISAC-MMD significantly outperformed the Nearest Neighbor, C4.5, Naïve Bayes, ALVOT, Naïve Associative Classifier, AIRS1, Immunos1, and CLONALG algorithms in conducting breast cancer classification.

Asunto(s)

RESUMEN

Today, society is more aware of their well-being and health, making wearable devices a new and affordable way to track them continuously. Smartwatches allow access to daily vital physiological measurements, which help people to be aware of their health status. Even though these technologies allow the following of different health conditions, their application in health is still limited to the following physical parameters to allow physicians treatment and diagnosis. This paper presents LM Research, a smart monitoring system mainly composed of a web page, REST APIs, machine learning algorithms, psychological questionnaire, and smartwatches. The system introduces the continuous monitoring of the users' physical and mental indicators to prevent a wellness crisis; the mental indicators and the system's continuous feedback to the user could be, in the future, a tool for medical specialists treating well-being. For this purpose, it collects psychological parameters on smartwatches and mental health data using a psychological questionnaire to develop a supervised machine learning wellness model that predicts the wellness of smartwatch users. The full construction of the database and the technology employed for its development is presented. Moreover, six machine learning algorithms (Decision Tree, Random Forest, Naive Bayes, Neural Networks, Support Vector Machine, and K-nearest neighbor) were applied to the database to test which classifies better the information obtained by the proposed system. In order to integrate this algorithm into LM Research, Random Forest being the one with the higher accuracy of 88%.

Asunto(s)

RESUMEN

Face recognition is a natural skill that a child performs from the first days of life; unfortunately, there are people with visual or neurological problems that prevent the individual from performing the process visually. This work describes a system that integrates Artificial Intelligence which learns the face of the people with whom the user interacts daily. During the study we propose a new hybrid model of Alpha-Beta Associative memories (Amαß) with Correlation Matrix (CM) and K-Nearest Neighbors (KNN), where the Amαß-CMKNN was trained with characteristic biometric vectors generated from images of faces from people who present different facial expressions such as happiness, surprise, anger and sadness. To test the performance of the hybrid model, two experiments that differ in the selection of parameters that characterize the face are conducted. The performance of the proposed model was tested in the databases CK+, CAS-PEAL-R1 and Face-MECS (own), which test the Amαß-CMKNN with faces of subjects of both sexes, different races, facial expressions, poses and environmental conditions. The hybrid model was able to remember 100% of all the faces learned during their training, while in the test in which faces are presented that have variations with respect to those learned the results range from 95.05% in controlled environments and 86.48% in real environments using the proposed integrated system.

Asunto(s)

RESUMEN

Classification of electromyographic signals has a wide range of applications, from clinical diagnosis of different muscular diseases to biomedical engineering, where their use as input for the control of prosthetic devices has become a hot topic of research. The challenge of classifying these signals relies on the accuracy of the proposed algorithm and the possibility of its implementation in hardware. This paper considers the problem of electromyography signal classification, solved with the proposed signal processing and feature extraction stages, with the focus lying on the signal model and time domain characteristics for better classification accuracy. The proposal considers a simple preprocessing technique that produces signals suitable for feature extraction and the Burg reflection coefficients to form learning and classification patterns. These coefficients yield a competitive classification rate compared to the time domain features used. Sometimes, the feature extraction from electromyographic signals has shown that the procedure can omit less useful traits for machine learning models. Using feature selection algorithms provides a higher classification performance with as few traits as possible. The algorithms achieved a high classification rate up to 100% with low pattern dimensionality, with other kinds of uncorrelated attributes for hand movement identification.