KEVIN LOPEZ - Academia.edu (original) (raw)
Papers by KEVIN LOPEZ
JAMIA Open
Objective Clinical guidelines recommend annual eye examinations to detect diabetic retinopathy (D... more Objective Clinical guidelines recommend annual eye examinations to detect diabetic retinopathy (DR) in patients with diabetes. However, timely DR detection remains a problem in medically underserved and under-resourced settings in the United States. Machine learning that identifies patients with latent/undiagnosed DR could help to address this problem. Materials and Methods Using electronic health record data from 40 631 unique diabetic patients seen at Los Angeles County Department of Health Services healthcare facilities between January 1, 2015 and December 31, 2017, we compared ten machine learning environments, including five classifier models, for assessing the presence or absence of DR. We also used data from a distinct set of 9300 diabetic patients seen between January 1, 2018 and December 31, 2018 as an external validation set. Results Following feature subset selection, the classifier with the best AUC on the external validation set was a deep neural network using majority ...
2018 XX Congreso Mexicano de Robótica (COMRob), 2018
This work presents the implementation of advanced time-delay controllers in a low-cost educationa... more This work presents the implementation of advanced time-delay controllers in a low-cost educational motion control platform based on a radio control (RC) servo. The highlights of the platform are presented as well as real-time experiments using the Proportional Integrated Retarded (PIR) and the Cascade Proportional Integrated Retarded (CPIR) algorithms. The experiments allow assessing the capabilities of the platform for implementing advanced controllers, and highlights the inherent filtering capabilities of time-delay controllers when they are fed by noisy measurements.
2017 4th International Conference on Control, Decision and Information Technologies (CoDIT), 2017
The goal of this work is to describe a new delay-based control law called the Cascade Proportiona... more The goal of this work is to describe a new delay-based control law called the Cascade Proportional Integral Retarded (CPIR) controller and its application to the position control of DC servodrives. The proposed controller has an inner loop-outer loop structure. The inner loop corresponds to an integral retarded (IR) controller that regulates the servodrive angular velocity. A proportional (P) controller closes the outer loop whose goal is to regulate the servodrive angular position. A tuning methodology for the CPIR controller is proposed and experiments using a laboratory prototype allow assessing its performance.
This work studies two delay-based nonlinear controllers and their tuning methods aimed to servodr... more This work studies two delay-based nonlinear controllers and their tuning methods aimed to servodrive control. The first controller is the Cascade Nonlinear Proportional Integral Retarded (CNPIR) control law, which is composed of two nested loops. The inner loop is regulated through an Integral Retarded (IR) Controller and the outer loop uses a nonlinear proportional (NP) controller. The second controller, which is called the Modified Nonlinear Proportional Integral Retarded (MNPIR) controller is a structural modification of the original Proportional Integral Retarded PIR controller, and also has an inner-outer loop structure. The topology of both controllers is presented as well as their tuning methodologies. Numerical results allow assessing the performance of these controllers when applied to a servodrive model.
Frontiers in Big Data, 2020
Choosing an optimal data fusion technique is essential when performing machine learning with mult... more Choosing an optimal data fusion technique is essential when performing machine learning with multimodal data. In this study, we examined deep learning-based multimodal fusion techniques for the combined classification of radiological images and associated text reports. In our analysis, we (1) compared the classification performance of three prototypical multimodal fusion techniques: Early, Late, and Model fusion, (2) assessed the performance of multimodal compared to unimodal learning; and finally (3) investigated the amount of labeled data needed by multimodal vs. unimodal models to yield comparable classification performance. Our experiments demonstrate the potential of multimodal fusion methods to yield competitive results using less training data (labeled data) than their unimodal counterparts. This was more pronounced using the Early and less so using the Model and Late fusion approaches. With increasing amount of training data, unimodal models achieved comparable results to multimodal models. Overall, our results suggest the potential of multimodal learning to decrease the need for labeled training data resulting in a lower annotation burden for domain experts.
IFAC-PapersOnLine, 2018
This work proposes a tuning methodology for the Cascade Proportional Integral Retarded (CPIR) con... more This work proposes a tuning methodology for the Cascade Proportional Integral Retarded (CPIR) controller applied to a class of second order systems. Unlike the Proportional Integral Retarded (PIR) control law, the CPIR is composed of two nested loops, each of them having its own controller. The inner loop is regulated through an Integral Retarded (IR) Controller and the outer loop uses a proportional (P) controller. It is worth remarking that the stability of both loops must be simultaneously guaranteed, a problem not appearing in the PIR controller. The proposed tuning methodology is based on the spectral abscissa analysis of the Outer loop. The IR controller tuning rules and a double root placement analysis in the CPIR quasy-polynomial allows writing an analytical form of the P proportional gain. Subsequently, this gain is computed by solving a transcendental nonlinear equation. The proposed tuning method is assessed by means of experiments in a laboratory prototype.
Journal of medical Internet research, May 19, 2017
The Canadian Computed Tomography (CT) Head Rule, a clinical decision rule designed to safely redu... more The Canadian Computed Tomography (CT) Head Rule, a clinical decision rule designed to safely reduce imaging in minor head injury, has been rigorously validated and implemented, and yet expected decreases in CT were unsuccessful. Recent work has identified empathic care as a key component in decreasing CT overuse. Health information technology can hinder the clinician-patient relationship. Patient-centered decision tools to support the clinician-patient relationship are needed to promote evidence-based decisions. Our objective is to formatively evaluate an electronic tool that not only helps clinicians at the bedside to determine the need for CT use based on the Canadian CT Head Rule but also promotes evidence-based conversations between patients and clinicians regarding patient-specific risk and patients' specific concerns. User-centered design with practice-based and participatory decision aid development was used to design, develop, and evaluate patient-centered decision suppo...
eGEMs (Generating Evidence & Methods to improve patient outcomes), 2015
and the Trauma Commons This Informatics Case Study is brought to you for free and open access by ... more and the Trauma Commons This Informatics Case Study is brought to you for free and open access by the the EDM Forum Products and Events at EDM Forum Community. It has been peer-reviewed and accepted for publication in eGEMs (Generating Evidence & Methods to improve patient outcomes).
Telemedicine and e-Health, 2013
Teleretinal screening with nonmydriatic cameras has been presented as a means of increasing the n... more Teleretinal screening with nonmydriatic cameras has been presented as a means of increasing the number of patients assessed for diabetic retinopathy in urban safety net clinics. It has been hypothesized that automated nonmydriatic cameras may improve screening rates by reducing the learning curve for camera use. In this article, we examine the impact of introducing automated nonmydriatic cameras to urban safety net clinics whose photographers had previously used manual cameras. We evaluated the impact of manual and automated digital nonmydriatic cameras on teleretinal screening using a quantitative analysis of readers' image quality ratings as well as a qualitative analysis, through in-depth interviews, of photographers' experiences. With the manual camera, 68% of images were rated "adequate" or better, including 24% rated "good" and 20% rated…
JAMIA Open
Objective Clinical guidelines recommend annual eye examinations to detect diabetic retinopathy (D... more Objective Clinical guidelines recommend annual eye examinations to detect diabetic retinopathy (DR) in patients with diabetes. However, timely DR detection remains a problem in medically underserved and under-resourced settings in the United States. Machine learning that identifies patients with latent/undiagnosed DR could help to address this problem. Materials and Methods Using electronic health record data from 40 631 unique diabetic patients seen at Los Angeles County Department of Health Services healthcare facilities between January 1, 2015 and December 31, 2017, we compared ten machine learning environments, including five classifier models, for assessing the presence or absence of DR. We also used data from a distinct set of 9300 diabetic patients seen between January 1, 2018 and December 31, 2018 as an external validation set. Results Following feature subset selection, the classifier with the best AUC on the external validation set was a deep neural network using majority ...
2018 XX Congreso Mexicano de Robótica (COMRob), 2018
This work presents the implementation of advanced time-delay controllers in a low-cost educationa... more This work presents the implementation of advanced time-delay controllers in a low-cost educational motion control platform based on a radio control (RC) servo. The highlights of the platform are presented as well as real-time experiments using the Proportional Integrated Retarded (PIR) and the Cascade Proportional Integrated Retarded (CPIR) algorithms. The experiments allow assessing the capabilities of the platform for implementing advanced controllers, and highlights the inherent filtering capabilities of time-delay controllers when they are fed by noisy measurements.
2017 4th International Conference on Control, Decision and Information Technologies (CoDIT), 2017
The goal of this work is to describe a new delay-based control law called the Cascade Proportiona... more The goal of this work is to describe a new delay-based control law called the Cascade Proportional Integral Retarded (CPIR) controller and its application to the position control of DC servodrives. The proposed controller has an inner loop-outer loop structure. The inner loop corresponds to an integral retarded (IR) controller that regulates the servodrive angular velocity. A proportional (P) controller closes the outer loop whose goal is to regulate the servodrive angular position. A tuning methodology for the CPIR controller is proposed and experiments using a laboratory prototype allow assessing its performance.
This work studies two delay-based nonlinear controllers and their tuning methods aimed to servodr... more This work studies two delay-based nonlinear controllers and their tuning methods aimed to servodrive control. The first controller is the Cascade Nonlinear Proportional Integral Retarded (CNPIR) control law, which is composed of two nested loops. The inner loop is regulated through an Integral Retarded (IR) Controller and the outer loop uses a nonlinear proportional (NP) controller. The second controller, which is called the Modified Nonlinear Proportional Integral Retarded (MNPIR) controller is a structural modification of the original Proportional Integral Retarded PIR controller, and also has an inner-outer loop structure. The topology of both controllers is presented as well as their tuning methodologies. Numerical results allow assessing the performance of these controllers when applied to a servodrive model.
Frontiers in Big Data, 2020
Choosing an optimal data fusion technique is essential when performing machine learning with mult... more Choosing an optimal data fusion technique is essential when performing machine learning with multimodal data. In this study, we examined deep learning-based multimodal fusion techniques for the combined classification of radiological images and associated text reports. In our analysis, we (1) compared the classification performance of three prototypical multimodal fusion techniques: Early, Late, and Model fusion, (2) assessed the performance of multimodal compared to unimodal learning; and finally (3) investigated the amount of labeled data needed by multimodal vs. unimodal models to yield comparable classification performance. Our experiments demonstrate the potential of multimodal fusion methods to yield competitive results using less training data (labeled data) than their unimodal counterparts. This was more pronounced using the Early and less so using the Model and Late fusion approaches. With increasing amount of training data, unimodal models achieved comparable results to multimodal models. Overall, our results suggest the potential of multimodal learning to decrease the need for labeled training data resulting in a lower annotation burden for domain experts.
IFAC-PapersOnLine, 2018
This work proposes a tuning methodology for the Cascade Proportional Integral Retarded (CPIR) con... more This work proposes a tuning methodology for the Cascade Proportional Integral Retarded (CPIR) controller applied to a class of second order systems. Unlike the Proportional Integral Retarded (PIR) control law, the CPIR is composed of two nested loops, each of them having its own controller. The inner loop is regulated through an Integral Retarded (IR) Controller and the outer loop uses a proportional (P) controller. It is worth remarking that the stability of both loops must be simultaneously guaranteed, a problem not appearing in the PIR controller. The proposed tuning methodology is based on the spectral abscissa analysis of the Outer loop. The IR controller tuning rules and a double root placement analysis in the CPIR quasy-polynomial allows writing an analytical form of the P proportional gain. Subsequently, this gain is computed by solving a transcendental nonlinear equation. The proposed tuning method is assessed by means of experiments in a laboratory prototype.
Journal of medical Internet research, May 19, 2017
The Canadian Computed Tomography (CT) Head Rule, a clinical decision rule designed to safely redu... more The Canadian Computed Tomography (CT) Head Rule, a clinical decision rule designed to safely reduce imaging in minor head injury, has been rigorously validated and implemented, and yet expected decreases in CT were unsuccessful. Recent work has identified empathic care as a key component in decreasing CT overuse. Health information technology can hinder the clinician-patient relationship. Patient-centered decision tools to support the clinician-patient relationship are needed to promote evidence-based decisions. Our objective is to formatively evaluate an electronic tool that not only helps clinicians at the bedside to determine the need for CT use based on the Canadian CT Head Rule but also promotes evidence-based conversations between patients and clinicians regarding patient-specific risk and patients' specific concerns. User-centered design with practice-based and participatory decision aid development was used to design, develop, and evaluate patient-centered decision suppo...
eGEMs (Generating Evidence & Methods to improve patient outcomes), 2015
and the Trauma Commons This Informatics Case Study is brought to you for free and open access by ... more and the Trauma Commons This Informatics Case Study is brought to you for free and open access by the the EDM Forum Products and Events at EDM Forum Community. It has been peer-reviewed and accepted for publication in eGEMs (Generating Evidence & Methods to improve patient outcomes).
Telemedicine and e-Health, 2013
Teleretinal screening with nonmydriatic cameras has been presented as a means of increasing the n... more Teleretinal screening with nonmydriatic cameras has been presented as a means of increasing the number of patients assessed for diabetic retinopathy in urban safety net clinics. It has been hypothesized that automated nonmydriatic cameras may improve screening rates by reducing the learning curve for camera use. In this article, we examine the impact of introducing automated nonmydriatic cameras to urban safety net clinics whose photographers had previously used manual cameras. We evaluated the impact of manual and automated digital nonmydriatic cameras on teleretinal screening using a quantitative analysis of readers' image quality ratings as well as a qualitative analysis, through in-depth interviews, of photographers' experiences. With the manual camera, 68% of images were rated "adequate" or better, including 24% rated "good" and 20% rated…