This work aims to classify malaria infected red blood cells from those uninfected using two deep learning approaches. Plasmodium parasite transmitted by a female anopheles’s mosquitoes bite is the main cause of malaria. Commonly, Microbiological analyses by a microscope allows detecting cells infected from a blood sample, followed by a specialist interpretation of results to conclude the diagnosis process. Taking advantage of efficient deep learning approaches applied in computer vision field, the present framework proposes two deep learning architecture based on Convolutional-Recurrent neural Networks to detect accurately malaria infected cells. The first one implements a Convolutional Long Short-Term Memory while the second uses a Convolutional Bidirectional Long Short-Term Memory architecture. A malaria’s public dataset consisting of parasitized and uninfected red blood cell images was used for training and testing the proposed models. The methods developed in this work achieved an accuracy of 99.89% in the detection of malaria-infected red blood cells, without preprocessing data.

There is a worldwide need for new methodologies to prediagnose breast cancer in an early stage, which helps to notably increase the possibility of saving the mammary gland or patient’s life. This work describes a new methodology proposal based on electrical impedance for the localization of preclinical carcinoma emulators in agar phantoms of the breast. The impedance is systematically measured through eight Ag/AgCl electrodes uniformly distributed in a ring arrangement placed on the breast agar phantom. The fundamental idea of the proposed location algorithm, named Anomaly Tracking Circle algorithm, is to find the breast agar area defined by straight lines joining the electrode pairs having the minimum difference value of the defined normalized impedance magnitude along the measurement sweep. Such difference is obtained with respect to a breast agar phantom without carcinoma emulator. The proposed methodology was evaluated through seven experimental agar models, six of them having carcinoma lobe emulators with different locations and electrical conductivities. According to the obtained results, the described methodology can obtain the location zone of preclinical-emulated carcinomas with an 83.33% success.

Malaria is a significant global health issue, especially in tropical regions. Accurate and rapid diagnosis is critical for effective treatment and reducing mortality rates. Traditional diagnostic methods, like blood smear microscopy, are time-intensive and prone to error. This study introduces a deep learning approach for classifying malaria-infected cells in blood smear images using convolutional neural networks (CNNs); Six CNN models were designed and trained using a large labeled dataset of malaria cell images, both infected and uninfected, and were implemented on the Jetson TX2 board to evaluate them. The model was optimized for feature extraction and classification accuracy, achieving 97.72% accuracy, and evaluated using precision, recall, and F1-score metrics and execution time. Results indicate deep learning significantly improves diagnostic time efficiency on embedded systems. This scalable, automated solution is particularly useful in resource-limited areas without access to expert microscopic analysis. Future work will focus on clinical validation.

This paper presents a comprehensive analysis of the effects of noise on the detection and localization of faults in transmission lines compensated with a unified power flow controller using traveling wave-based methods. This study focuses on the impact of harmonic and transient noises, which are inherent to power generation, transmission, and UPFC operation. A novel algorithm is proposed combining the Discrete Wavelet Transform and Clarke Transform to detect and localize faults under various noise conditions. The algorithm is tested on a simulated transmission line model in MATLAB/Simulink (Version R2022b) with noise levels of 20 dB, 30 dB, and 40 dB and transient frequencies of 1 kHz, 5 kHz, and 10 kHz. The results demonstrate that the algorithm achieves an average fault localization error of 0.523% under harmonic noise and 0.777% under transient noise, with fault detection rates of 96.3% and 90.75%, respectively. This study highlights the robustness of the traveling wave method in noisy environments and provides insights into the challenges posed by UPFC-compensated lines.

A multifrequency electrical impedance tomography (EIT) system was evaluated for its ability to detect conductive inclusions simulating carcinomas in breast phantoms, with a comparative analysis of time-difference (TD) and frequency-difference (FD) reconstruction approaches. The proposed EIT V5 system employs two concentric rings of 16 electrodes to acquire surface voltage measurements at multiple excitation frequencies (50 kHz, 500 kHz, and 1 MHz). Image reconstruction was performed using the linear back projection algorithm, and system performance was quantitatively assessed through spatial overlap metrics (intersection over union, IoU, and F 1-score), contrast-to-noise ratio (CNR), and confusion-matrix-derived metrics (sensitivity, specificity, and precision). The area under the receiver operating characteristic curve (AUC) was also computed as a pixel-level, threshold-independent separability metric. Experimental phantoms were designed to approximate breast tissue composition, consisting primarily of adipose material with embedded conductive inclusions representing tumors. The results show that FD EIT consistently outperforms TD reconstruction across all evaluated scenarios, achieving higher CNR values (>2.4) and improved spatial agreement (IoU and F 1-score), while TD reconstructions exhibit significant variability and reduced contrast at lower frequencies. Although high AUC values (>0.99) are observed for the FD approach, these should be interpreted as indicators of conductivity separability within individual reconstructions rather than diagnostic performance. Reconstructions obtained from the lower electrode ring demonstrate increased sensitivity, highlighting the influence of electrode geometry and inclusion proximity on detection performance. Importantly, FD reconstructions enhance contrast between conductive inclusions and surrounding tissue without requiring a prior baseline measurement. These findings indicate that multifrequency, FD EIT provides a robust and reliable approach for detecting conductive anomalies in controlled phantom conditions, reducing reconstruction artifacts and improving tissue discrimination. The proposed methodology shows strong potential as an auxiliary tool for early breast cancer detection, intracranial hemorrhage monitoring, and the development of wearable biomedical imaging systems.

Introducción: La medición y caracterización de la impedancia eléctrica de los tejidos biológicos permite distinguir entre tejidos sanos y patológicos. El uso de técnicas como la tomografía de impedancia eléctrica es muy utilizada en pruebas clínicas a través de diversas aplicaciones. En este trabajo se emplea el método Linear Back Proyection (LBP) para reconstruir imágenes a partir de la medición de impedancia eléctrica con el objetivo de localizar emuladores de carcinoma insertados en modelos de agar-agar con forma de mama femenina.Método: Un arreglo anillar de ocho electrodos fue empleado para realizar mediciones de impedancia en siete modelos de agar-agar con forma de mama femenina. Se diseñó y simuló en el software COMSOL un corte transversal de una mama femenina sana (sin emuladores) que representa el plano de medición del arreglo anillar de electrodos. Se calcularon los mapas de sensibilidad de cada par de electrodos a lo largo de la medición de impedancia utilizando los datos de la distribución de conductividad del modelo simulado. Finalmente, la reconstrucción de la imagen de tomografía eléctrica para cada modelo se obtuvo aplicando el algoritmo LBP a los valores de impedancia eléctrica medidos en el arreglo anillar de electrodos.Resultados: Se calcularon los mapas de sensibilidad a partir de las mediciones de impedancia obtenidas de cada modelo experimental considerado. Como resultado se generó una matriz de sensibilidad que caracteriza la distribución de los potenciales eléctricos dentro del plano de medición simulado en COMSOL. La zona de localización obtenida fue correcta para todos los modelos experimentales con emuladores de carcinomas insertados fuera del centro del arreglo anillar de electrodos.Discusión o Conclusión: De acuerdo con los resultados obtenidos del proyecto descrito, la metodología propuesta permite la generación de imágenes para localizar emuladores de carcinoma con diámetros aproximados de 1 cm, ello a través de un enfoque no iterativo. Por lo anterior, la localización no requiere una alta complejidad computacional. La localización de un emulador de carcinoma insertado en la zona central del modelo de mama femenina no se obtuvo correctamente debido a la baja sensibilidad de la configuración anillar de los electrodos en dicha zona.