Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 13
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Automatic Quality Electrogram Assessment Improves Reentrant Activity Identification in Atrial Fibrillation

  • Costoya-Sánchez, Alejandro
  • Hernández-Romero, Ismael
  • Liberos, Alejandro
  • Fernández-Avilés, Francisco
  • Narayan, Sanjiv M.
  • Atienza, Felipe
  • Rodrigo, Miguel
  • Andreu M. Climent|||0000-0002-7260-8811
  • Guillem Sánchez, María Salud|||0000-0001-5660-3693
[EN] Location of reentrant electrical activity responsible for driving atrial fibrillation (AF) is key to ablative therapies. The aim of this work is to study the effect of the quality of the electrograms (EGMs) used for 3D phase analysis on reentrant activity identification, as well as to develop an algorithm capable of automatically identifying lowquality signals. EGMs signals from 259 episodes obtained from 29 AF patients were recorded using 64-electrode basket catheters. Low-quality EGMs were manually identified. Reentrant activity was identified in 3D phase maps and provided an area under the ROC curve (AUC) of 0.69 when compared to a 2D activation-based method. Reentries located in regions with poor-quality EGMs were then removed, increasing the AUC to 0.80. The EGM classification algorithm showed a similar performance both for low-quality EGM identification (sensitivity 0.91 and specificity 0.80) and for reentrant activity location with 3D phase analysis (AUC 0.80). Discard of reentrant activity identified in regions where EGMs showed low quality significantly improved the specificity of the 3D phase analysis. Besides, EGMs classification according to their quality proved to be possible using time and spectral domain parameters., This work was supported by the Instituto de Salud Carlos III FEDER (DTS16/00160; PI16/01123; PI17/01059; PI17/01106), the EIT-Health 19600 AFFINE and the Valencian Regional Government (AICO2018).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Non-Invasive Electrophysiological Mapping Entropy Predicts Atrial Fibrillation Ablation Efficacy Better Than Clinical Characteristics

  • De la Nava, A. S.
  • Fabregat, M. C.
  • Rodrigo, M.
  • Hernández, I.
  • Liberos, A.
  • Fernández-Avilés, F.
  • Atienza, F.
  • Guillem Sánchez, María Salud|||0000-0001-5660-3693
  • Andreu M. Climent|||0000-0002-7260-8811
[EN] Success rate of atrial fibrillation (AF) ablation remains far from satisfactory. In this study, a 6 months AF freedom predictive model based on Fuzzy Entropy of non-invasive body surface potential maps is compared with clinical predictors. The study included 29 patients referred for pulmonary vein isolated catheter ablation procedure. Non-invasive electrocardiographic mapping with 54 ECG electrodes was recorded for all patients during the ablation procedure. Six months follow up was used to evaluate the efficacy of the ablation procedure. Predictions based on non-invasive electrocardiographic mappings during adenosine infusion (accuracy: 90%, AUC: 0.93) showed a clear improvement over standard-of-care clinical parameter models (accuracy: 62.1%, AUC: 0. 54). Our results indicate that measurements of electrophysiological complexity of AF signals could improve the clinical practice by predicting the efficacy of AF ablation procedures., This work was supported by the Instituto de Salud Carlos III FEDER (DTS16/00160; PI16/01123; PI17/01059; PI17/01106; EIT-Health 19600 AFFINE)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Temporal Stability of Dominant Frequency as Predictor of Atrial Fibrillation Recurrence

  • Costoya-Sánchez, Alejandro
  • Hernández-Romero, Ismael
  • Liberos, Alejandro
  • Fernández-Avilés, Francisco
  • Narayan, Sanjiv M.
  • Atienza, Felipe
  • Rodrigo, Miguel
  • Andreu M. Climent|||0000-0002-7260-8811
  • Guillem Sánchez, María Salud|||0000-0001-5660-3693
[EN] Catheter ablation is one of the main therapies for restoring sinus rhythm in patients with atrial fibrillation (AF), yet AF termination ratios are far from satisfactory. The goal of this work is to study if temporal stability of dominant frequencies (DFs) of electrograms (EGMs) can be used as predictor of AF recurrence. EGMs were recorded from 29 AF patients using 64- pole basket catheters during the ablation procedure. DFs before ablation were obtained for 4-second overlapping fragments of EGM recordings with a 0.4 s shift, and their temporal stability was evaluated for short-term (between 8 and 12 s) and long-term time intervals (>5 min). Patients were classified as AF (N=15) if sinus rhythm was not maintained in a 12-month post-ablation followup, and AF free otherwise (N=14). Significant differences were found in the short-term analysis between AF free and AF patients for the difference between the mode value in DFs (p=0.045), as well as for the long-term analysis for the normalized average between DFs (p=0.028) and the average between DFs (p=0.043). More stable values were found for AF free patients for all statistically significant metrics. Short- and long-term temporal stability of DF values of EGM signals were found to be associated with the 12- month success rate of ablative therapies of AF patients., This work was supported by the Instituto de Salud Carlos III FEDER (DTS16/00160; PI16/01123; PI17/01059; PI17/01106), the EIT-HEALTH 19600 AFFINE and the Valencian Regional Government (AICO2018).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Detection of Atrial Fibrillation Driver Locations Using CNN and Body Surface Potentials

  • Cámara-Vázquez, Miguel Ángel
  • Hernández-Romero, Ismael
  • Morgado-Reyes, Eduardo
  • Barquero-Pérez, Óscar
  • Guillem Sánchez, María Salud|||0000-0001-5660-3693
  • Andreu M. Climent|||0000-0002-7260-8811
[EN] Atrial fibrillation (AF) is characterized by complex and irregular propagation patterns, and AF onset locations and drivers responsible for its perpetuation are main targets for ablation procedures. Several Deep Learningbased methods have proposed to detect AF, but the estimation of the atrial area where the drivers are found is a topic where further research is needed. In this work, we propose to estimate the zone where AF drivers are found from body surface potentials (BSPs) and Convolutional Neural Networks (CNN), modeling a supervised classification problem. Accuracy in the test set was 0.89 when using noisy BSPs (SNR=20dB), while the Cohen¿s Kappa was 0.85. Therefore, the proposed method could help to identify target regions for ablation using a non-invasive procedure, and avoiding the use of ECG Imaging (ECGI)., This work has been partially supported by: Ministerio de Ciencia e Innovacion (PID2019-105032GB-I00), Instituto de Salud Carlos III, and Ministerio de Ciencia, Innovacion y Universidades (supported by FEDER Fondo Europeo de Desarrollo Regional PI17/01106 and RYC2018-024346B-750), Consejeria de Ciencia, Universidades e Innovacion of the Comunidad de Madrid through the program RIS3 (S-2020/L2-622), EIT Health (Activity code 19600, EIT Health is supported by EIT, a body of the European Union) and the European Union's Horizon 2020 research and innovation program under the Marie Skodowska-Curie grant agreement No. 860974.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Contribución de congreso (ConferenceOutput)

Effects of Geometry in Atrial Fibrillation Markers Obtained With Electrocardiographic Imaging

  • Molero-Alabau, Rubén
  • Hernández-Romero, Ismael
  • Liberos, Alejandro
  • Fernández-Avilés, Francisco
  • Atienza, Felipe
  • Rodrigo, Miguel
  • Andreu M. Climent|||0000-0002-7260-8811
  • Guillem Sánchez, María Salud|||0000-0001-5660-3693
[EN] Electrocardiographic imaging (ECGI) can characterise cardiac pathologies such as atrial fibrillation (AF) through specific markers based on frequency or phase analysis. In this study, the effect of the geometry of patients torso and atria in the ECGI resolution is studied. A realistic 3D atrial geometry was located on 30 patient torsos and ECGI signals were calculated for 30 different AF simulations in each torso. Dominant frequency (DF) and reentrant activity analysis were calculated for each scenario. Anatomical and geometrical measurements of each torso (30-80% of variability between patients) and atria were calculated and compared with the errors in the ECGI estimation versus the departing EGM maps. Results show evidences that big chest dimensions worsen the non-invasive calculation of AF markers (p<0.05). Also, higher number of visible electrodes from each atrial region improves ECGI characterization measured as lower DF deviations (0.64±0.26 Hz vs 0.72±0.27 Hz, p<0.05) and higher reentrant activity coincidence (10.1±12.2% vs 3.4±3.4%, p<0.05). Torso and atrial geometry affect the quality of the non-invasive reconstruction of AF markers such as DF or reentrant activity. Knowing the geometrical parameters that worsen non-invasive AF maps may help to measure each detected AF driver reliability., Supported in part by: Instituto de Salud Carlos III FEDER (Fondo Europeo de Desarrollo Regional; IJCI2014-22178, DTS16/00160; PI14/00857, PI16/01123; PI17/01059; PI17/01106), Generalitat Valenciana Grants (APOSTD/2017 and APOSTD/2018) and projects (GVA/2018/103) and EIT-Health 19600 AFFINE.
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

Development of an Open Source Tool and a Multi-Platform for Generation of Forensic Reports

  • Marti, Mireia
  • Vila Avendaño, Pilar
  • Vázquez-Naya, José
[Abstract] Computer Forensics is a science that is part of computer security and focuses on identifying, preserving, analyzing and presenting electronic evidence that has been found on a device. This process has to be thoroughly documented by the expert who carries it out, and must be adapted to standards such as UNE 197010:2015 or ISO/IEC 27042:2015. However, there are no tools to facilitate this task. Therefore, in this work, a multiplatform and open source tool is developed to facilitate the expert’s elaboration of the report, and the management of the documentation related to the case, while keeping this information safe., This work was supported by the “Collaborative Project in Genomic Data Integration (CICLOGEN)” PI17/01826 funded by the Carlos III Health Institute from the Spanish National plan for Scientific and Technical Research and Innovation 2013-2016 and the European Regional Development Funds (FEDER)—“A way to build Europe”. This project was also supported by the General Directorate of Culture, Education and University Management of Xunta de Galicia ED431D 2017/16 and “Drug Discovery Galician Network” Ref. ED431G/01 and the “Galician Network for Colorectal Cancer Research” (Ref. ED431D 2017/23), and finally by the Spanish Ministry of Economy and Competitiveness for its support through the funding of the unique installation BIOCAI (UNLC08-1E-002, UNLC13-13-3503) and the European Regional Development Funds (FEDER) by the European Union. Additional support was offered by the Consolidation and Structuring of Competitive Research Units—Competitive Reference Groups (ED431C 2018/49), funded by the Ministry of Education, University and Vocational Training of the Xunta de Galicia endowed with EU FEDER funds, Xunta de Galicia; ED431D 2017/16, Xunta de Galicia; ED431G/01, Xunta de Galicia; ED431D 2017/23, Xunta de Galicia; ED431C 2018/49
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

Identification of Prevotella, Anaerotruncus and Eubacterium Genera by Machine Learning Analysis of Metagenomic Profiles for Stratification of Patients Affected by Type I Diabetes

  • Fernández-Edreira, Diego
  • Liñares Blanco, José
  • Fernández-Lozano, Carlos
[Abstract] Previous works have reported different bacterial strains and genera as the cause of different clinical pathological conditions. In our approach, using the fecal metagenomic profiles of newborns, a machine learning-based model was generated capable of discerning between patients affected by type I diabetes and controls. Furthermore, a random forest algorithm achieved a 0.915 in AUROC. The automation of processes and support to clinical decision making under metagenomic variables of interest may result in lower experimental costs in the diagnosis of complex diseases of high prevalence worldwide., This work was supported by the “Collaborative Project in Genomic Data Integration (CICLOGEN)” PI17/01826 funded by the Carlos III Health Institute from the Spanish National plan for Scientific and Technical Research and Innovation 2013–2016 and the European Regional Development Funds (FEDER)—“A way to build Europe.” and the General Directorate of Culture, Education and University Management of Xunta de Galicia (Ref. ED431G/01, ED431D 2017/16), the “Galician Network for Colorectal Cancer Research” (Ref. ED431D 2017/23) and Competitive Reference Groups (Ref. ED431C 2018/49). The funding body did not have a role in the experimental design; data collection, analysis and interpretation; and writing of this manuscript, Xunta de Galicia; ED431G/01, Xunta de Galicia; ED431D 2017/16, Xunta de Galicia; ED431D 2017/23, Xunta de Galicia; ED431C 2018/49
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

Retina-Focused Preprocessing for Enhanced Deep Learning Classification of Multiple Sclerosis from OCT Scans

  • Álvarez-Rodríguez, Lorena
  • Cordón Ciordia, Beatriz
  • García-Martín, Elena
  • Novo Buján, Jorge
  • Moura, Joaquim de
  • Ortega Hortas, Marcos
Presentado en: EUROCAST 2026, 20th International Conference on Computer Aided Systems Theory Las Palmas de Gran Canaria, Spain, February 23- 27, 2026, [Abstract]: This work researches the impact of retina-focused preprocessing on OCT-based deep learning classification of Multiple Sclerosis (MS). Raw OCT B-scans were compared with anatomically preprocessed images, where the retina was segmented, flattened, and isolated from background structures. Experiments with several convolutional architectures showed that focusing on the retinal region enhances classification accuracy and consistency across repetitions, supporting the relevance of anatomically guided preprocessing in OCT-based neurological screening., This work was supported by the Instituto de Salud Carlos III (ISCIII), Government of Spain [grant numbers PI17/01726, PI20/00437, PI23/00935, RD21/0007/0022 (Inflammatory Disease Network- RICORS), FORT23/00010 (Programa FORTALECE)]; MICIU/AEI/10.13039/501100011033 and by “ERDF A way of making Europe” [grant numbers PID2023-148913OB-I00, PID2024-161024OB-I00]; the Consellería de Educación, Universidade, e Formación Profesional, Xunta de Galicia, Grupos de Referencia Competitiva [grant number ED431C 2024/33]; and by the Government of Aragon [group B23_23R and PROY_B50_24].
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

Clinical Decision Support Tool for the Identification of Pathological Structures Associated with Age-Related Macular Degeneration

  • Barrientos Lema, Iván
  • Moura, Joaquim de
  • Novo Buján, Jorge
  • Ortega Hortas, Marcos
  • Penedo, Manuel
18th International Conference on Computer Aided Systems Theory, EUROCAST 2022, Las Palmas de Gran Canaria, 20 - 25 February 2022, This version of the article has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature's AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/978-3-031-25312-6_48, [Abstract]: In the field of ophthalmology, different imaging modalities are commonly used to carry out different clinical diagnostic procedures. Currently, both optical coherence tomography (OCT) and optical coherence tomography angiography (OCT-A) have made great advances in the study of the posterior pole of the eye and are essential for the diagnosis and monitoring of the treatment of different ocular and systemic diseases. On the other hand, the development of clinical decision support systems is an emerging field, in which clinical and technological advances are allowing clinical specialists to diagnose various pathologies with greater precision, which translates into more appropriate treatment and, consequently, an improvement in the quality of life of patients. This paper presents a clinical decision support tool for the identification of different pathological structures associated with age-related macular degeneration using OCT and OCT-A images. The system provides a useful tool that facilitates clinical decision-making in the diagnosis and treatment of this relevant disease., This research was funded by Instituto de Salud Carlos III, Government of Spain, DTS18/00136 and PI17/00940 research projects; Ministerio de Ciencia e Innovación y Universidades, Government of Spain, RTI2018-095894-B-I00 research project; Ministerio de Ciencia e Innovación, Government of Spain through the research project with reference PID2019-108435RB-I00; Consellería de Cultura, Educación e Universidade, Xunta de Galicia, Grupos de Referencia Competitiva, grant ref. ED431C 2020/24 and postdoctoral grant ref. ED481B 2021/059; Axencia Galega de Innovación (GAIN), Xunta de Galicia, grant ref. IN845D 2020/38; CITIC, Centro de Investigación de Galicia ref. ED431G 2019/01, receives financial support from Con-sellería de Educación, Universidade e Formación Profesional, Xunta de Galicia, through the ERDF (80%) and Secretaría Xeral de Universidades (20%)., Xunta de Galicia; ED431C 2020/24, Xunta de Galicia; ED481B 2021/059, Xunta de Galicia; IN845D 2020/38, Xunta de Galicia; ED431G 2019/01
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

Transformer-Based Multi-Prototype Approach for Diabetic Macular Edema Analysis in OCT Images

  • Vidal, Plácido
  • Moura, Joaquim de
  • Novo Buján, Jorge
  • Ortega Hortas, Marcos
  • Cardoso, Jaime S.
© 2023 IEEE. This version of the paper has been accepted for publication. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The final published paper is available online at: https://doi.org/10.1109/ICASSP49357.2023.10095039, Presentado en: ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Rhodes Island, Greece, 04-10 June 2023, [Abstract]: Optical Coherence Tomography (OCT) is the major diagnostic tool for the leading cause of blindness in developed countries: Diabetic Macular Edema (DME). Depending on the type of fluid accumulations, different treatments are needed. In particular, Cystoid Macular Edemas (CMEs) represent the most severe scenario, while Diffuse Retinal Thickening (DRT) is an early indicator of the disease but a challenging scenario to detect. While methodologies exist, their explanatory power is limited to the input sample itself. However, due to the complexity of these accumulations, this may not be enough for a clinician to assess the validity of the classification. Thus, in this work, we propose a novel approach based on multi-prototype networks with vision transformers to obtain an example-based explainable classification. Our proposal achieved robust results in two representative OCT devices, with a mean accuracy of 0.9099 ± 0.0083 and 0.8582 ± 0.0126 for CME and DRT-type fluid accumulations, respectively., This research was funded by Instituto de Salud Carlos III, Government of Spain, PI17/00940 research project; Ministerio de Ciencia e Innovación y Universidades, Government of Spain, RTI2018-095894-B-I00 research project, Ayudas para la formación de profesorado universitario (FPU), grant ref. FPU18/02271 and Ayudas complementarias de movilidad destinadas a beneficiarios del programa de Formación del Profesorado Universitario (FPU) EST22/00696; Ministerio de Ciencia e Innovación, Government of Spain through the research project with reference PID2019-108435RB-I00; Consellería de Cultura, Educación e Universidade, Xunta de Galicia, Grupos de Referencia Competitiva, grant ref. ED431C 2020/24; Axencia Galega de Innovación (GAIN), Xunta de Galicia, grant ref. IN845D 2020/38;CITIC, Centro de Investigación de Galicia ref. ED431G 2019/01, receives financial support from Consellería de Educación, Universidade e Formación Profesional, Xunta de Galicia, through the ERDF (80%) and Secretaría Xeral de Universidades (20%)., Xunta de Galicia; ED431C 2020/24, Xunta de Galicia; IN845D 2020/38, Xunta de Galicia; ED431G 2019/01
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

Sistema automático para la evaluación de enfermedades neurodegenerativas en imágenes OCT mediante deep learning

  • Álvarez-Rodríguez, Lorena
  • Pueyo-Bestué, Ana
  • Moura, Joaquim de
  • Viladés, Elisa
  • García-Martín, Elena
  • Sánchez, Clara I.
  • Novo Buján, Jorge
  • Ortega Hortas, Marcos
O IDIS, INIBIC e IIS Galicia Sur organizaron Biointegrasaúde 2023, celebrado no Complexo Hospitalario Universitario de A Coruña (CHUAC), 14 de dicembro de 2023., El Alzhéimer (AD), el temblor esencial (ET), la esclerósis múltiple (MS) o el Párkinson (PD) son enfermedades neurodegenerativas (END) que están correlacionadas con cambios en algunas capas retinales clave. Las tomografías de coherencia óptica (OCT) pueden proporcionar información detallada desde diferentes perspectivas para analizar esas capas. Proponemos un sistema automático que: Segmenta las capas RNFL y GCL-BM de vólumenes 3D e imágenes OCT retinales de mácula y disco óptico. Extrae biomarcadores computacionales como deep features o grosor. Hace un cribado combinando información de ambas vistas OCT., Esta investigación ha sido subvencionada por el Instituto de Salud Carlos III, Gobierno de Spain, [proyectos de investigación PI17/01726 y PI20/00437]; Inflammatory Disease Network (RICORS) [proyecto RD21/0002/0050]; Ministerio de Ciencia e Innovación, Gobierno de Spain [proyectos de investigación PDC2022-133132-I00, PID2019-108435RB-I00 y TED2021-131201B-I00]; Consellería de Cultura, Educación e Universidade, Xunta de Galicia, a través de Grupos de Referencia Competitiva [subvención ED431C 2020/24]. Las organizaciones financiadoras no participaron en el dise˜no o ejecución de esta investigación., Gobierno de Aragón; RD21/0002/0050, Xunta de Galicia; ED431C 2020/24
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

Geometric Deep Learning for Essential Tremor Screening Using

  • Álvarez-Rodríguez, Lorena
  • Moura, Joaquim de
  • Viladés, Elisa
  • García-Martín, Elena
  • Novo Buján, Jorge
  • Ortega Hortas, Marcos
This version of the article has been accepted for publication, after peer review. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The Version of Record is available online at: https://doi.org/10.1109/IJCNN64981.2025.11229335 Traballo presentado en: 2025 International Joint Conference on Neural Networks (IJCNN), Rome, Italy, 30 June - 05 July 2025, [Abstract]: Essential tremor (ET) is a prevalent movement disorder characterized by motor and non-motor symptoms, often associated with neurodegeneration. Optical coherence tomography (OCT) has emerged as a valuable tool to identify retinal biomarkers in ET patients. This study presents a novel methodology for ET detection using 3D point clouds derived from retinal OCT layers. Leveraging advanced geometric deep learning (GDL) architectures, including PointTransformer, PointCNN, PointNet++ and SplineCNN, we evaluated the diagnostic potential of individual retinal layers, including the Retinal Nerve Fiber Layer (RNFL), Ganglion Cell Layer (GCL) and Bruch’s Membrane (BM), as well as their combined representation. Our approach achieved state-of-the-art results, with PointTransformer obtaining an F1-score of 0.85 using only BM retinal surface, while requiring just 2% of the original point cloud size. These findings underscore the diagnostic value of OCT-derived 3D data and demonstrate the potential of GDL for computational biomarker extraction in neurodegenerative disorders, offering a scalable and efficient framework for ET diagnosis., This work was supported by the Instituto de Salud Carlos III (ISCIII), Government of Spain [grant numbers PI17/01726, PI20/00437, PI23/00935, RD21/0002/0050 (Inflammatory Disease Network - RICORS), FORT23/00010 (Programa FORTALECE)], the Ministerio de Ciencia e Innovacion, Government of Spain [grant numbers PID2023-148913OB-I00, TED2021-131201B-I00, and PDC2022-133132-I00], the Consellería de Educacion, Universidade, e Formación Profesional, Xunta de Galicia, Grupos de Referencia Competitiva [grant number ED431C 2024/33] and by the Government of Aragon [group B2323R]. This work was also supported by the Horizon Europe Programme through the ACHILLES/101189689 project (HORIZON-CL4-2024-DATA-01-01)., Xunta de Galicia; ED431C 2024/33
RUC. Repositorio da Universidade da Coruña
Publicación Contribución de congreso (ConferenceOutput)

3D Point Cloud Analysis via Transformer-Based Graph Learning for Multiple Sclerosis Screening in OCT Images

  • Álvarez-Rodríguez, Lorena
  • García Prego, Iván
  • Moura, Joaquim de
  • Pueyo-Bestué, Ana
  • Viladés, Elisa
  • García-Martín, Elena
  • Sánchez, Clara I.
  • Novo Buján, Jorge
  • Ortega Hortas, Marcos
Presented at: 28th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems (KES 2024), [Abstract]: Multiple Sclerosis (MS), the leading cause of non-traumatic neurological impairment in young adults, manifests morphological changes in the retina observable in Optical Coherence Tomography (OCT) images. These changes in the Retinal Nerve Fibre Layer (RNFL) and the Ganglion Cell Layer - Bruch’s Membrane (GCL-BM) serve as potential computational biomarkers for MS. In this work, we propose a transformer-based graph learning approach for analyzing 3D point clouds generated from RNFL and GCL-BM contours, marking a first in the application of geometric deep learning (GDL) to MS diagnosis via OCT scans. Our proposal, tailored for efficiency, synergizes the global contextual strengths of transformers with the detailed, structure-aware capabilities of graph neural networks. Such integration allows for the nuanced analysis of complex retinal structures, significantly boosting the precision of MS detection by uncovering patterns not discernible to the human eye. Additionally, we conducted a comprehensive study on the optimal downsampling size of input 3D point clouds, ensuring efficient data processing without compromising diagnostic accuracy. Our optimal configuration achieved a test F1-Score of 0.88, using only 4.0% of total 3D points, showcasing the effectiveness of our method despite the higher computational demands compared to less complex, albeit less precise, configurations. These promising results are the first in the study of 3D analysis and transformer-based geometric deep learning for MS screening based on OCT images, which are revolutionizing neurophtalmological research., Funded by Instituto de Salud Carlos III (ISCIII), Government of Spain, [projects PI17/01726 and PI20/00437]; Inflammatory Disease Network (RICORS) [project RD21/0002/0050]; Ministerio de Ciencia e Innovacion, Government of Spain [ projects PID2019-108435RB-I00, TED2021-131201B-I00, and PDC2022-133132-I00]; Consellería de Educacion, Universidade, e Formación Profesional, Xunta de Galicia, Grupos de Referencia Competitiva, [grant number ED431C 2020/24]. Also supported by the ISCIII under the grant [FORT23/00010] as part of the Programa FORTALECE of Ministerio de Ciencia e Innovacion., Xunta de Galicia; ED431C 2020/24