Pasar al contenido principal

APOYO A LA DECISION EN OFTALMOLOGIA BASADO EN MACHINE LEARNING Y APLICADO A IMAGENES MULTI-MODALES DE LA RETINA

PID2022-143299OB-I00

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2022
Plan Estatal 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 6
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Semi-supervised fuzzy DBN-based broad learning system for forecasting ICU admissions in post-transplant COVID-19 patients

  • Zhang, Xiao
  • Nebot Castells, M. Àngela|||0000-0002-4621-8262
This paper introduces a novel semi-supervised neuro-fuzzy system to predict ICU admissions among post-COVID organ transplant recipients. Addressing the challenges of small sample sizes and lacking labels in organ transplantation, our study takes on these issues by proposing a DBN-Based Dual Manifold Regularized Fuzzy Broad Learning System (D-DMR-FBLS). This system utilizes the streamlined and flat architecture of the Broad Learning System (BLS), integrating Deep Belief Networks (DBN) and Takagi-Sugeno-Kang (TSK) systems to enhance representation learning capacities during the Unsupervised Training Phase (UTP). The system combines the strong feature learning capabilities of DBN with the powerful fuzzy rule extraction capacity of the TSK system, enhancing the model’s predictive performance and generalization capability. Moreover, we propose two types of graph-based manifold regularization, sample-based and feature-based, within this novel D-DMR-FBLS framework. Our method enhances its predictive ability by exploiting both the similarity among unlabeled and labeled patient samples, as well as the correlations between features within the fuzzy feature space. Employed to predict ICU admission risks in post-transplant COVID-19 patients, the method has demonstrated superior performance over existing methods, particularly in scenarios with limited samples and labels, thereby providing more accurate decision support for medical professionals in optimizing resource allocation for transplant patients., This paper is part of project PID2022-143299OB-I00, financed by MCIN/AEI/10.13030/501100011033/FEDER,UE., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Semi-supervised dual-manifold regularized fuzzy broad learning for ICU admission prediction in post-COVID transplant recipients

  • Zhang, Xiao
  • Nebot Castells, M. Àngela|||0000-0002-4621-8262
This paper introduces a novel semi-supervised Dual-Manifold Regularized Fuzzy Broad Learning System (DMR-FBLS) designed for predicting intensive care unit (ICU) admissions among post-COVID organ transplant recipients. Addressing the challenge of increased mortality risk and strained medical resources among patients who are COVID-19-infected transplant recipients, our study fills the gap by employing machine learning to predict the ICU requirements for this particular group of patients. The DMR-FBLS uniquely integrates two manifold regularizations - feature and sample manifold regularization - into the fuzzy broad learning system (FBLS) framework. This integration enhances the system's capacity to utilize the latent structural relationships within the data. Evaluated using the IDOTCOVID database, the DMR-FBLS demonstrated superior performance over other models in terms of accuracy, F1-score, and G-mean. In scenarios with limited sample availability, DMR-FBLS outperformed both the FBLS and TSK fuzzy systems in terms of accuracy and G-mean. This underscores the efficacy of DMR-FBLS as a semi-supervised learning fuzzy system, accurately predicting ICU demands and thereby providing decision support for medical professionals in optimizing resource allocation for transplant patients., This research has been supported by the project PID2022-143299OB-I00., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Diabetic retinopathy prediction from OCTA-based vessel tortuosity metrics using machine learning

  • Guijarro Heeb, Torben
  • Zarranz Ventura, Javier
  • Romero Merino, Enrique|||0000-0003-2404-5716
  • Vellido Alcacena, Alfredo|||0000-0002-9843-1911
Optical Coherence Tomography Angiography (OCTA) has emerged in the last decade as a reference imaging technology that offers a non-invasive, rapid method for the detailed assessment of microvascular changes at the capillary level. In this brief study, we assess, using statistical and Machine Learning models, the usefulness of a vascular tortuosity-based OCTA image transformation for the diagnostic discrimination between Type 1 Diabetes Mellitus patients with mild or no presence of Diabetic Retinopathy (DR), and those with advanced DR., This study was funded by Spanish research grant PID2022-143299OB-I00/AEI/10.13039/501100011033/FEDER, UE., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Semi-supervised fuzzy DBN-based broad learning system for forecasting ICU admissions in post-transplant COVID-19 patients

  • Zhang, Xiao
  • Nebot Castells, M. Àngela
This paper introduces a novel semi-supervised neuro-fuzzy system to predict ICU admissions among post-COVID organ transplant recipients. Addressing the challenges of small sample sizes and lacking labels in organ transplantation, our study takes on these issues by proposing a DBN-Based Dual Manifold Regularized Fuzzy Broad Learning System (D-DMR-FBLS). This system utilizes the streamlined and flat architecture of the Broad Learning System (BLS), integrating Deep Belief Networks (DBN) and Takagi-Sugeno-Kang (TSK) systems to enhance representation learning capacities during the Unsupervised Training Phase (UTP). The system combines the strong feature learning capabilities of DBN with the powerful fuzzy rule extraction capacity of the TSK system, enhancing the model’s predictive performance and generalization capability. Moreover, we propose two types of graph-based manifold regularization, sample-based and feature-based, within this novel D-DMR-FBLS framework. Our method enhances its predictive ability by exploiting both the similarity among unlabeled and labeled patient samples, as well as the correlations between features within the fuzzy feature space. Employed to predict ICU admission risks in post-transplant COVID-19 patients, the method has demonstrated superior performance over existing methods, particularly in scenarios with limited samples and labels, thereby providing more accurate decision support for medical professionals in optimizing resource allocation for transplant patients., This paper is part of project PID2022-143299OB-I00, financed by MCIN/AEI/10.13030/501100011033/FEDER,UE., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Diabetic retinopathy prediction from OCTA-based vessel tortuosity metrics using machine learning

  • Guijarro Heeb, Torben
  • Zarranz Ventura, Javier
  • Romero Merino, Enrique
  • Vellido Alcacena, Alfredo
Optical Coherence Tomography Angiography (OCTA) has emerged in the last decade as a reference imaging technology that offers a non-invasive, rapid method for the detailed assessment of microvascular changes at the capillary level. In this brief study, we assess, using statistical and Machine Learning models, the usefulness of a vascular tortuosity-based OCTA image transformation for the diagnostic discrimination between Type 1 Diabetes Mellitus patients with mild or no presence of Diabetic Retinopathy (DR), and those with advanced DR., This study was funded by Spanish research grant PID2022-143299OB-I00/AEI/10.13039/501100011033/FEDER, UE., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Semi-supervised dual-manifold regularized fuzzy broad learning for ICU admission prediction in post-COVID transplant recipients

  • Zhang, Xiao
  • Nebot Castells, M. Àngela
This paper introduces a novel semi-supervised Dual-Manifold Regularized Fuzzy Broad Learning System (DMR-FBLS) designed for predicting intensive care unit (ICU) admissions among post-COVID organ transplant recipients. Addressing the challenge of increased mortality risk and strained medical resources among patients who are COVID-19-infected transplant recipients, our study fills the gap by employing machine learning to predict the ICU requirements for this particular group of patients. The DMR-FBLS uniquely integrates two manifold regularizations - feature and sample manifold regularization - into the fuzzy broad learning system (FBLS) framework. This integration enhances the system's capacity to utilize the latent structural relationships within the data. Evaluated using the IDOTCOVID database, the DMR-FBLS demonstrated superior performance over other models in terms of accuracy, F1-score, and G-mean. In scenarios with limited sample availability, DMR-FBLS outperformed both the FBLS and TSK fuzzy systems in terms of accuracy and G-mean. This underscores the efficacy of DMR-FBLS as a semi-supervised learning fuzzy system, accurately predicting ICU demands and thereby providing decision support for medical professionals in optimizing resource allocation for transplant patients., This research has been supported by the project PID2022-143299OB-I00., Peer Reviewed, Postprint (author's final draft)