Jorge Rendulich
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jorge Rendulich has a total of 36 co-authors. Here are the top co-authors by number of shared publications:
- Erasmo Sulla Espinoza
Universidad Nacional de San Agustín de Arequipa, Perú
7 shared publications - Jorge Rodolfo Beingolea Garay
Department of Electronic Systems Engineering, University of São Paulo, São Paulo 05508-010, Brazil
6 shared publications - Milagros Zegarra
School of Electronic Engineering, Universidad Nacional de San Agustin de Arequipa, Arequipa 04000, Peru
6 shared publications - Carlos Esteban Mamani Huisa
Universidad Nacional de San Agustín de Arequipa
2 shared publications - Edward Carello Figueroa Tejada
Universidad Nacional de San Agustín de Arequipa, Perú
2 shared publications - Edwin M. Chambi
School of Electronic Engineering, Universidad Nacional de San Agustin de Arequipa, Arequipa 04000, Peru
2 shared publications
Latest Publications
- Open Access
Performance Evaluation of LoRaWAN for Monitoring People with Disabilities at University Campus
The growing need to foster inclusive education in university environments has driven the development of technological solutions aimed at improving the academic experiences of students with disabilities. These individuals often face barriers to autonomy and participation, especially on large and complex campuses. This article presents the performance evaluation of a LoRaWAN network specifically designed for monitoring people with disabilities on a university campus. The system aims to provide equitable access to campus resources and real-time support to students with disabilities. Leveraging the advantages of Low-Power Wide-Area Networks (LPWAN), particularly LoRaWAN, the proposed system enables real-time tracking with broad coverage and minimal power consumption, without requiring any active user interaction. Each student receives a wearable LoRa-enabled device that wirelessly communicates with a network of gateways strategically installed throughout the campus. To evaluate the system’s performance, this work conducts link-level experiments focusing on the communication between the LoRa end devices (nodes) and the central gateway. The analysis focuses on the network coverage, signal strength (RSSI), signal-to-noise ratio (SNR), and packet reception rate (PRR). The experimental results confirmed that the proposed system is technically robust and operationally effective under real campus conditions. Beyond its technical contributions, the proposed solution represents a concrete step toward building safer and more accessible academic environments that reinforce the autonomy and inclusion of students with disabilities.
Open AccessPerformance Evaluation of LoRaWAN for Monitoring People with Disabilities at University Campus
The growing need to foster inclusive education in university environments has driven the development of technological solutions aimed at improving the academic experiences of students with disabilities. These individuals often face barriers to autonomy and participation, especially on large and complex campuses. This article presents the performance evaluation of a LoRaWAN network specifically designed for monitoring people with disabilities on a university campus. The system aims to provide equitable access to campus resources and real-time support to students with disabilities. Leveraging the advantages of Low-Power Wide-Area Networks (LPWAN), particularly LoRaWAN, the proposed system enables real-time tracking with broad coverage and minimal power consumption, without requiring any active user interaction. Each student receives a wearable LoRa-enabled device that wirelessly communicates with a network of gateways strategically installed throughout the campus. To evaluate the system’s performance, this work conducts link-level experiments focusing on the communication between the LoRa end devices (nodes) and the central gateway. The analysis focuses on the network coverage, signal strength (RSSI), signal-to-noise ratio (SNR), and packet reception rate (PRR). The experimental results confirmed that the proposed system is technically robust and operationally effective under real campus conditions. Beyond its technical contributions, the proposed solution represents a concrete step toward building safer and more accessible academic environments that reinforce the autonomy and inclusion of students with disabilities.
- Open Access
- 1 January 2025
- Published by Revista Mexicana de Ingenieria Biomedica in Revista Mexicana de Ingeniería Biomédica
- Vol. 46 (1) , e1490
Algoritmo de Delineación Especifico basado en el EWMA de dos Etapas para la Estimación de la Presión Arterial no Invasiva
El ruido dinámico fuerte, debido a su variabilidad e intensidad, impide que los métodos convencionales de detección de picos en señales ECG y PPG basados en umbrales fijos funcionen correctamente en dispositivos portátiles. La inflexibilidad de estos umbrales fijos resulta en una baja sensibilidad y valor predictivo positivo. Por lo tanto, en este trabajo se propone un algoritmo de delineación específico con umbral adaptativo basado en el Modelo de Suavizado Exponencial Ponderado (EWMA) de dos etapas, enfocándose en la flexibilidad, precisión, robustez frente al ruido dinámico fuerte y baja carga computacional. El algoritmo propuesto demostró un desempeño robusto en condiciones de alto SNR (24 dB y 18 dB), alcanzando una sensibilidad y un valor predictivo positivo del 100 %. En condiciones de ruido moderado (12 dB), el algoritmo mantuvo una alta sensibilidad del 99.39 % y un valor predictivo positivo del 98.18 %, con una tasa de error de delineación (DER) del 2.43 %. Incluso en condiciones de bajo SNR (6 dB), el algoritmo superó significativamente a los enfoques basados en umbrales fijos, en comparación con más del 50 % en métodos convencionales. Además, se validó el algoritmo utilizando un modelo matemático para estimar la presión arterial basado en el tiempo de tránsito del pulso, con señales provenientes de la base de datos MIMIC. Los resultados mostraron un error medio de -1.422 mmHg para la presión arterial sistólica (SBP) y 0.577 mmHg para la presión arterial diastólica (DBP), con desviaciones estándar de 4.668 mmHg y 2.888 mmHg, respectivamente, cumpliendo con los estándares de la Asociación para el Avance de la Instrumentación Médica (AAMI).
- 1 January 2025
- Published by Revista Mexicana de Ingenieria Biomedica in Revista Mexicana de Ingeniería Biomédica
- Vol. 46 (1) , e1490
Open AccessAlgoritmo de Delineación Especifico basado en el EWMA de dos Etapas para la Estimación de la Presión Arterial no Invasiva
El ruido dinámico fuerte, debido a su variabilidad e intensidad, impide que los métodos convencionales de detección de picos en señales ECG y PPG basados en umbrales fijos funcionen correctamente en dispositivos portátiles. La inflexibilidad de estos umbrales fijos resulta en una baja sensibilidad y valor predictivo positivo. Por lo tanto, en este trabajo se propone un algoritmo de delineación específico con umbral adaptativo basado en el Modelo de Suavizado Exponencial Ponderado (EWMA) de dos etapas, enfocándose en la flexibilidad, precisión, robustez frente al ruido dinámico fuerte y baja carga computacional. El algoritmo propuesto demostró un desempeño robusto en condiciones de alto SNR (24 dB y 18 dB), alcanzando una sensibilidad y un valor predictivo positivo del 100 %. En condiciones de ruido moderado (12 dB), el algoritmo mantuvo una alta sensibilidad del 99.39 % y un valor predictivo positivo del 98.18 %, con una tasa de error de delineación (DER) del 2.43 %. Incluso en condiciones de bajo SNR (6 dB), el algoritmo superó significativamente a los enfoques basados en umbrales fijos, en comparación con más del 50 % en métodos convencionales. Además, se validó el algoritmo utilizando un modelo matemático para estimar la presión arterial basado en el tiempo de tránsito del pulso, con señales provenientes de la base de datos MIMIC. Los resultados mostraron un error medio de -1.422 mmHg para la presión arterial sistólica (SBP) y 0.577 mmHg para la presión arterial diastólica (DBP), con desviaciones estándar de 4.668 mmHg y 2.888 mmHg, respectivamente, cumpliendo con los estándares de la Asociación para el Avance de la Instrumentación Médica (AAMI).
- Open Access
LoRa Technology Enhanced with a Custom-Designed High-Gain Yagi-Uda Antenna for Data Transmission from Misti Volcano Monitoring to Arequipa City
This study details the design and implementation of a high-gain Yagi-Uda antenna network for the transmission of real-time monitoring data from the Misti Volcano to the city of Arequipa. As Misti is classified as a high-risk volcano due to its active volcanic nature and the close proximity of nearly one million inhabitants, the current monitoring infrastructure is insufficient to meet the demands of effective surveillance. In response, this project integrates Internet of Things (IoT) technology, the LoRa (Long Range) network, and an optimized seven-element Yagi-Uda antenna, developed using advanced optimization algorithms to enhance transmission efficiency. The primary objective is to facilitate the reliable collection and transmission of critical sensor data for subsequent analysis by volcanological experts, thereby supporting improved prediction and mitigation of potential volcanic hazards. Field tests have demonstrated that the Yagi-Uda antenna, when coupled with LoRa technology, achieved uninterrupted data transmission over a distance of 16 km. The integration of IoT, LoRa, and the optimized antenna design offers a scalable and resilient solution for the continuous monitoring and risk assessment of Misti, enabling the incorporation of advanced high-precision sensors for enhanced surveillance capabilities.
Open AccessLoRa Technology Enhanced with a Custom-Designed High-Gain Yagi-Uda Antenna for Data Transmission from Misti Volcano Monitoring to Arequipa City
This study details the design and implementation of a high-gain Yagi-Uda antenna network for the transmission of real-time monitoring data from the Misti Volcano to the city of Arequipa. As Misti is classified as a high-risk volcano due to its active volcanic nature and the close proximity of nearly one million inhabitants, the current monitoring infrastructure is insufficient to meet the demands of effective surveillance. In response, this project integrates Internet of Things (IoT) technology, the LoRa (Long Range) network, and an optimized seven-element Yagi-Uda antenna, developed using advanced optimization algorithms to enhance transmission efficiency. The primary objective is to facilitate the reliable collection and transmission of critical sensor data for subsequent analysis by volcanological experts, thereby supporting improved prediction and mitigation of potential volcanic hazards. Field tests have demonstrated that the Yagi-Uda antenna, when coupled with LoRa technology, achieved uninterrupted data transmission over a distance of 16 km. The integration of IoT, LoRa, and the optimized antenna design offers a scalable and resilient solution for the continuous monitoring and risk assessment of Misti, enabling the incorporation of advanced high-precision sensors for enhanced surveillance capabilities.
- Open Access
- 1 September 2024
- Published by Revista Mexicana de Ingenieria Biomedica in Revista Mexicana de Ingeniería Biomédica
- Vol. 45 (3) , 112-128
Development of a Frequent Exchange Coordinate Algorithm for Detection of Precordial Electrode Exchange during ECG based on Error and Correlation Parameters
In order to develop a method for detecting precordial electrode exchange, a frequent exchange coordinate (FEC) algorithm was implemented, that identified the minimum correlation points necessary to detect a specific electrode exchange, and its performance was tested with error estimators (mean square error and percent mean square difference). Validation of the algorithm was performed using the k-fold cross-validation technique on the PTB, Chapman University and Shaoxing Hospital, PTB XL, and Georgia 12-lead ECG Challenge databases. The results indicate averages of Se= 99.16 % and Sp= 99.38 % (MSE), Se= 95.38 % and Sp= 99.47 % (PRD), Se= 98.44 % and Sp= 99.49 % (Pearson), Se= 98.45 % and Sp= 99.48 % (modified Pearson), Se= 95.39 % and Sp= 99.81 % (Bray Curtis), Se= 80.00 % and Sp= 97.84 % (correlation sign). MSE presents a significant improvement in execution time (61.49µs N=1000), representing, on average, 44.99 % of the execution time for Pearson correlation. The frequent exchange coordinates algorithm was then validated using signal analysis with the mean square error (MSE), representing a good alternative to detect electrode exchange in real time, easy to implement, and low computational cost.
- 1 September 2024
- Published by Revista Mexicana de Ingenieria Biomedica in Revista Mexicana de Ingeniería Biomédica
- Vol. 45 (3) , 112-128
Open AccessDevelopment of a Frequent Exchange Coordinate Algorithm for Detection of Precordial Electrode Exchange during ECG based on Error and Correlation Parameters
In order to develop a method for detecting precordial electrode exchange, a frequent exchange coordinate (FEC) algorithm was implemented, that identified the minimum correlation points necessary to detect a specific electrode exchange, and its performance was tested with error estimators (mean square error and percent mean square difference). Validation of the algorithm was performed using the k-fold cross-validation technique on the PTB, Chapman University and Shaoxing Hospital, PTB XL, and Georgia 12-lead ECG Challenge databases. The results indicate averages of Se= 99.16 % and Sp= 99.38 % (MSE), Se= 95.38 % and Sp= 99.47 % (PRD), Se= 98.44 % and Sp= 99.49 % (Pearson), Se= 98.45 % and Sp= 99.48 % (modified Pearson), Se= 95.39 % and Sp= 99.81 % (Bray Curtis), Se= 80.00 % and Sp= 97.84 % (correlation sign). MSE presents a significant improvement in execution time (61.49µs N=1000), representing, on average, 44.99 % of the execution time for Pearson correlation. The frequent exchange coordinates algorithm was then validated using signal analysis with the mean square error (MSE), representing a good alternative to detect electrode exchange in real time, easy to implement, and low computational cost.
- Open Access
- 24 July 2024
- Published by MDPI AG in Electronics
- Vol. 13 (15) , 2912
Benchmarking Time-Frequency Representations of Phonocardiogram Signals for Classification of Valvular Heart Diseases Using Deep Features and Machine Learning
Heart sounds and murmur provide crucial diagnosis information for valvular heart diseases (VHD). A phonocardiogram (PCG) combined with modern digital processing techniques provides a complementary tool for clinicians. This article proposes a benchmark different time–frequency representations, which are spectograms, mel-spectograms and cochleagrams for obtaining images, in addition to the use of two interpolation techniques to improve the quality of the images, which are bicubic and Lanczos. Deep features are extracted from a pretrained model called VGG16, and for feature reduction, the Boruta algorithm is applied. To evaluate the models and obtain more precise results, nested cross-validation is used. The best results achieved in this study were for the cochleagram with 99.2% accuracy and mel-spectogram representation with the bicubic interpolation technique, which reached 99.4% accuracy, both having a support vector machine (SVM) as a classifier algorithm. Overall, this study highlights the potential of time–frequency representations of PCG signals combined with modern digital processing techniques and machine learning algorithms for accurate diagnosis of VHD.
- 24 July 2024
- Published by MDPI AG in Electronics
- Vol. 13 (15) , 2912
Open AccessBenchmarking Time-Frequency Representations of Phonocardiogram Signals for Classification of Valvular Heart Diseases Using Deep Features and Machine Learning
Heart sounds and murmur provide crucial diagnosis information for valvular heart diseases (VHD). A phonocardiogram (PCG) combined with modern digital processing techniques provides a complementary tool for clinicians. This article proposes a benchmark different time–frequency representations, which are spectograms, mel-spectograms and cochleagrams for obtaining images, in addition to the use of two interpolation techniques to improve the quality of the images, which are bicubic and Lanczos. Deep features are extracted from a pretrained model called VGG16, and for feature reduction, the Boruta algorithm is applied. To evaluate the models and obtain more precise results, nested cross-validation is used. The best results achieved in this study were for the cochleagram with 99.2% accuracy and mel-spectogram representation with the bicubic interpolation technique, which reached 99.4% accuracy, both having a support vector machine (SVM) as a classifier algorithm. Overall, this study highlights the potential of time–frequency representations of PCG signals combined with modern digital processing techniques and machine learning algorithms for accurate diagnosis of VHD.
- Open Access
- 13 May 2024
- Published by MDPI AG in Preprints.org
Benchmarking Time-Frequency Representations of PCG Signals for Classification Valvular Heart Diseases Using Deep Features and Machine Learning
Heart sounds and murmurprovidecrucial diagnosis information for valvular heart diseases (VHD). Phonocardiogram (PCG) combined with modern digital processing techniques, provides a complementary tool for clinicians. This article proposes a benchmark different time-frequency representations, which are spectogram, mel-spectogram and cochleagram for obtaining images, in addition to the use of two interpolation techniques to improve the quality of the images, which are Bicubic and Lanczos. Deep features are extracted from a pre-trained model called VGG16 and for feature reduction the Boruta algorithm is applied. To evaluate the models and obtain more precise results, nested cross-validation is used. The best results achieved in this study were for coclegram with 99.2% accuracy and mel-spectogram representation with the bicubic interpolation technique which reached 99.4% accuracy, both having support vector machine (SVM) as classifier 10 algorithm. Overall, this study highlights the potential of time-frequency representations of PCG 11 signals combined with modern digital processing techniques and machine learning algorithms for 12 accurate diagnosis of VHD.
- 13 May 2024
- Published by MDPI AG in Preprints.org
Open AccessBenchmarking Time-Frequency Representations of PCG Signals for Classification Valvular Heart Diseases Using Deep Features and Machine Learning
Heart sounds and murmurprovidecrucial diagnosis information for valvular heart diseases (VHD). Phonocardiogram (PCG) combined with modern digital processing techniques, provides a complementary tool for clinicians. This article proposes a benchmark different time-frequency representations, which are spectogram, mel-spectogram and cochleagram for obtaining images, in addition to the use of two interpolation techniques to improve the quality of the images, which are Bicubic and Lanczos. Deep features are extracted from a pre-trained model called VGG16 and for feature reduction the Boruta algorithm is applied. To evaluate the models and obtain more precise results, nested cross-validation is used. The best results achieved in this study were for coclegram with 99.2% accuracy and mel-spectogram representation with the bicubic interpolation technique which reached 99.4% accuracy, both having support vector machine (SVM) as classifier 10 algorithm. Overall, this study highlights the potential of time-frequency representations of PCG 11 signals combined with modern digital processing techniques and machine learning algorithms for 12 accurate diagnosis of VHD.
- Open Access
- 30 March 2024
- Published by Politechnika Lubelska in Applied Computer Science
- Vol. 20 (1) , 157-172
COMPARISON AND EVALUATION OF LMS-DERIVED ALGORITHMS APPLIED ON ECG SIGNALS CONTAMINATED WITH MOTION ARTIFACT DURING PHYSICAL ACTIVITIES
The acquisition of ECG signals offers physicians and specialists a very important tool in the diagnosis of cardiovascular diseases. However, very often these signals are affected by noise from various sources, including noise generated by movement during physical activity. This type of noise is known as Motion Artifact (MA) which changes the waveform of the signal, leading to erroneous readings. The elimination of this noise is performed by different filtering techniques, where the adaptive filtering using the LMS (least mean squares) algorithm stands out. The objective of this article is to determine which algorithms best deal with motion artifacts, taking into account the use of instruments or wearable equipment, in different conditions of physical activity. A comparison between different algorithms derived from LMS (NLMS, PNLMS and IPNLM) used in adaptive filtering is carried out using indicators such as: Pearson's Correlation Coefficient, Signal to Noise Ratio (SNR) and Mean Squared Error (MSE) as metrics to evaluate them. For this purpose, the mHealth database was used, which contains ECG signals taken during moderate to medium intensity physical activities. The results show that filtering by IPNLMS as well as PNLMS offers an improvement both visually and in terms of SNR, Pearson, and MSE indicators.
- 30 March 2024
- Published by Politechnika Lubelska in Applied Computer Science
- Vol. 20 (1) , 157-172
Open AccessCOMPARISON AND EVALUATION OF LMS-DERIVED ALGORITHMS APPLIED ON ECG SIGNALS CONTAMINATED WITH MOTION ARTIFACT DURING PHYSICAL ACTIVITIES
The acquisition of ECG signals offers physicians and specialists a very important tool in the diagnosis of cardiovascular diseases. However, very often these signals are affected by noise from various sources, including noise generated by movement during physical activity. This type of noise is known as Motion Artifact (MA) which changes the waveform of the signal, leading to erroneous readings. The elimination of this noise is performed by different filtering techniques, where the adaptive filtering using the LMS (least mean squares) algorithm stands out. The objective of this article is to determine which algorithms best deal with motion artifacts, taking into account the use of instruments or wearable equipment, in different conditions of physical activity. A comparison between different algorithms derived from LMS (NLMS, PNLMS and IPNLM) used in adaptive filtering is carried out using indicators such as: Pearson's Correlation Coefficient, Signal to Noise Ratio (SNR) and Mean Squared Error (MSE) as metrics to evaluate them. For this purpose, the mHealth database was used, which contains ECG signals taken during moderate to medium intensity physical activities. The results show that filtering by IPNLMS as well as PNLMS offers an improvement both visually and in terms of SNR, Pearson, and MSE indicators.
- 26 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Heterogeneous Devices: Network Layer Integration Experience
The work proposes the development of an integration architecture for highly heterogeneous sensor network ecosystems. The implementation is carried out on a device called the “integration device”. The device functions as a management and abstraction layer, integrated with the communication and data layer of a service-oriented middleware. The integration device controls real-time events through the programming of thread groups, which have the role of managing and abstracting the heterogeneity of data and communication protocols of wireless sensor devices. A part of the integration device implementation is presented, and data transfer rate experiments are conducted to measure its performance.
- 26 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Heterogeneous Devices: Network Layer Integration Experience
The work proposes the development of an integration architecture for highly heterogeneous sensor network ecosystems. The implementation is carried out on a device called the “integration device”. The device functions as a management and abstraction layer, integrated with the communication and data layer of a service-oriented middleware. The integration device controls real-time events through the programming of thread groups, which have the role of managing and abstracting the heterogeneity of data and communication protocols of wireless sensor devices. A part of the integration device implementation is presented, and data transfer rate experiments are conducted to measure its performance.
- 25 March 2023
- Published by Springer Nature
- p. 803-811
PCG Heart Sounds Quality Classification Using Neural Networks and SMOTE Tomek Links for the Think Health Project
- 25 March 2023
- Published by Springer Nature
- p. 803-811
PCG Heart Sounds Quality Classification Using Neural Networks and SMOTE Tomek Links for the Think Health Project
- Open Access
Assistive Devices: Technology Development for the Visually Impaired
Technology has been contributing significantly to the development of assistive devices for disabled persons (DPs). Many of these devices aim to assist people who are blind or visually impaired, providing them with friendlier ways to interact with their surroundings (obstacles, objects, and navigation). However, the high cost of these devices makes it difficult for DPs to purchase them. The development of an assistive device kit to be used by the visually impaired in controlled environments (indoor) or urban spaces (outdoor) is presented in this work with a didactic and detailed approach. This low-cost kit (USD 50 per device) consists of a Smart Cane, Smart Cap, and Smart Glove. The methodology addressed aims to support and encourage the reproduction, customization, and manufacture of new assistive devices with low cost and wide applicability. The devices are functional, benefit usability, and, due to the characteristics of the project and the materials used in their construction, do not impact the ergonomics of the visually impaired or disabled person who may use these devices. The devices that integrate the assistive kit can be used independently or combined. In addition to having detection, fall-control, navigation, and real-time tracking functions, they are independent of additional local infrastructure for their use.
Open AccessAssistive Devices: Technology Development for the Visually Impaired
Technology has been contributing significantly to the development of assistive devices for disabled persons (DPs). Many of these devices aim to assist people who are blind or visually impaired, providing them with friendlier ways to interact with their surroundings (obstacles, objects, and navigation). However, the high cost of these devices makes it difficult for DPs to purchase them. The development of an assistive device kit to be used by the visually impaired in controlled environments (indoor) or urban spaces (outdoor) is presented in this work with a didactic and detailed approach. This low-cost kit (USD 50 per device) consists of a Smart Cane, Smart Cap, and Smart Glove. The methodology addressed aims to support and encourage the reproduction, customization, and manufacture of new assistive devices with low cost and wide applicability. The devices are functional, benefit usability, and, due to the characteristics of the project and the materials used in their construction, do not impact the ergonomics of the visually impaired or disabled person who may use these devices. The devices that integrate the assistive kit can be used independently or combined. In addition to having detection, fall-control, navigation, and real-time tracking functions, they are independent of additional local infrastructure for their use.