Pasar al contenido principal

DESARROLLO DE UN SOFTWARE DE ANALISIS DE IMAGEN DE RM CEREBRAL MULTIMODAL DE ULTRA-ALTA RESOLUCION PARA SU APLICACION A ENTORNOS CLINICOS

PID2023-152127OB-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 2023
Plan Estatal 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITÈCNICA DE VALÈNCIA

Resultados relacionados

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

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

DeepCERES: A deep learning method for cerebellar lobule segmentation using ultra-high resolution multimodal MRI

  • Morell-Ortega, Sergio|||0009-0001-7639-2915
  • Vivó, Roberto|||0000-0002-0751-4114
  • Rubio Navarro, Gregorio
  • DE LA IGLESIA VAYÁ, MARIA DE LOS DESAMPARADOS
  • Manjón Herrera, José Vicente|||0000-0001-6640-927X
  • Ruiz-Perez, Marina
  • Gadea, Marien
  • Aparici-Robles, Fernando
  • Catheline, Gwenaelle
  • Mansecal, Boris
  • Coupé, Pierrick
[EN] This paper introduces a novel multimodal and high-resolution human brain cerebellum lobule segmentation method. Unlike current tools that operate at standard resolution (1 mm3 ) or using mono-modal data, the proposed method improves cerebellum lobule segmentation through the use of a multimodal and ultra-high resolution (0.125 mm3 ) training dataset. To develop the method, first, a database of semi-automatically labelled cerebellum lobules was created to train the proposed method with ultra-high resolution T1 and T2 MR images. Then, an ensemble of deep networks has been designed and developed, allowing the proposed method to excel in the complex cerebellum lobule segmentation task, improving precision while being memory efficient. Notably, our approach deviates from the traditional U-Net model by exploring alternative architectures. We have also integrated deep learning with classical machine learning methods incorporating a priori knowledge from multi-atlas segmentation which improved precision and robustness. Finally, a new online pipeline, named DeepCERES, has been developed to make available the proposed method to the scientific community requiring as input only a single T1 MR image at standard resolution., This work has been developed thanks to the project PID2020-118608RB-I00 and to the project PID2023-152127OB-I00 of the Ministerio de Ciencia, Innovacion y Universidades de España. This work also benefited from the support of the project DeepvolBrain and HoliBrain of the French National Research Agency (ANR-18-CE45-0013 and ANR23-CE45-0020-01). This study was achieved within the Laboratory of Excellence TRAIL ANR-10-LABX-57 for the BigDataBrain project. Moreover, we thank the Investments for the future Program IdEx Bordeaux (ANR-10- IDEX- 03- 02, HL-MRI Project), Cluster of excellence CPU and the CNRS.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Ultra-high resolution multimodal MRI densely labelled holistic structural brain atlas

  • Manjón Herrera, José Vicente|||0000-0001-6640-927X
  • Morell-Ortega, Sergio|||0009-0001-7639-2915
  • Rubio Navarro, Gregorio
  • Ruiz-Perez, Marina
  • Mansecal, Boris
  • Le Bot, Edern
  • Gadea, Marien
  • Lanuza, Enrique
  • Catheline, Gwenaelle
  • Tourdias, Thomas
  • Planche, Vicent
  • Giraud, Rémi
  • Rivière, Denis
  • Mangin, Jean-Francois
  • Labra-Avila, Nicole
  • Vivo-Hernando, Roberto
  • Aparici-Robles, Fernando
  • de la Iglesia-Vaya, María
  • Coupé, Pierrick
[EN] In this paper, we introduce a novel structural holistic Atlas (holiAtlas) of human brain anatomy based on multimodal and high-resolution MRI that covers several anatomical levels from the organ level to the substructure level, using a new protocol for dense labelling generated from the fusion of multiple local protocols at different scales. This atlas was constructed by averaging images and segmentations of 75 healthy subjects from the Human Connectome Project database. Specifically, 3T MR images of T1, T2 and WMn (White Matter nulled) contrasts at 0.125 mm3 resolution were selected for this project. The images of these 75 subjects were nonlinearly registered and averaged using symmetric group-wise normalisation to construct the atlas. At the finest level, the proposed atlas has 350 different labels derived from 7 distinct delineation protocols. These labels were grouped at multiple scales, offering a coherent and consistent holistic representation of the brain across different levels of detail. This multiscale and multimodal atlas can be used to develop new ultra-high-resolution segmentation methods, potentially improving the early detection of neurological disorders. We make it publicly available to the scientific community., This work has been developed thanks to the projects PID2020-118608RB-I00 and PID2023-152127OB-I00 of the Ministerio de Ciencia e Innovacion de España. This work also benefited from the support of the projects DeepvolBrain, HoliBrain and FOLDDICO of the French National Research Agency (ANR-18-CE45-0013, ANR-23-CE45-0020-01 and ANR-20-CHIA-0027- 01). Finally, this study received financial support from the French government in the framework of the University of Bordeaux's France 2030 program / RRI "IMPACT, the PEPR StratifyAging and the IHU VBHI (ANR-23-IAHU-0001). We thank the support of ITACA (Institute of Information and Communication Technologies) at UPV (Universitat Politècnica de València).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

petBrain: a new pipeline for amyloid, Tau tangles and neurodegeneration quantification using PET and MRI

  • Coupé, Pierrick
  • Mansencal, Boris
  • Morandat, Floréal
  • Villain, Nicolas
  • Planche, Vincent
  • Morell-Ortega, Sergio|||0009-0001-7639-2915
  • Manjón Herrera, José Vicente|||0000-0001-6640-927X
[EN] Introduction Quantification of amyloid plaques (A), neurofibrillary tangles (T2), and neurodegeneration (N) using PET and MRI is critical for Alzheimer¿s disease (AD) diagnosis and prognosis. Existing pipelines face limitations regarding processing time, tracer variability handling, and multimodal integration. Methods We developed petBrain, a novel end-to-end processing pipeline for amyloid-PET, tau-PET, and structural MRI. It leverages deep learning-based segmentation, standardized biomarker quantification (Centiloid, CenTauR, HAVAs), and simultaneous estimation of A, T2, and N biomarkers. It is implemented in a web-based format, requiring no local computational infrastructure and software usage knowledge. Results petBrain provides reliable, rapid quantification with results comparable to existing pipelines for A and T2, showing strong concordance with data processed in ADNI databases. The staging and quantification of A/T2/N by petBrain demonstrated good agreements with CSF/plasma biomarkers, clinical status and cognitive performance. Discussion petBrain represents a powerful open platform for standardized AD biomarker analysis, facilitating clinical research applications., This work benefited from the support of the project HoliBrain of the French National Research Agency (ANR-23-CE45-0020-01). Moreover, this project is supported by the Precision and global vascular brain health institute funded by the France 2030 investment plan as part of the IHU3 initiative (ANR-23- IAHU- 0001). Finally, this study received financial support from the French government in the framework of the University of Bordeaux s France 2030 program / RRI IMPACT and the PEPR StratifyAging and Prodrom-ND. This work benefited from the support of the project ChatvolBrain from CNRS. This work was also granted access to the HPC resources of IDRIS under the allocation 2022-AD011013848R1 made by GENCI. This work also benefited from the support of the project PID2023-152127OB-I00 of the Ministerio de Ciencia, Innovacion y Universidades of Spain.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Interthalamic adhesion alterations in multiple sclerosis: associations with thalamic damage and cognition

  • Fournet, Arthur
  • Munsch, Fanny
  • Koubiyr, Ismail
  • Vidal, Julie P.
  • Saubusse, Aurore
  • Charre-Morin, Julie
  • Lataste, Katy K.
  • Lavielle, Audrey
  • Bei, Zhang
  • Coupé, Pierrick
  • Barbeau, Emmanuel J.
  • Kumar, Vinod J.
  • Hornberger, Michael
  • Dousset, Vincent
  • Morell-Ortega, Sergio|||0009-0001-7639-2915
  • Manjón Herrera, José Vicente|||0000-0001-6640-927X
[EN] Background/objectives We investigated whether the interthalamic adhesion (IA), a midline structure connecting the thalami, is altered in multiple sclerosis (MS) and associated with thalamic damages and cognition. Methods We prospectively included 32 clinically isolated syndrome/early MS, 31 relapsing-remitting MS, 31 primary-progressive MS patients, and 103 matched controls. All underwent anatomical 3 T MRI and completed a comprehensive cognitive battery. IA presence, subtype, and volume were assessed by two blinded readers. Thalamic nuclei and other brain structures were segmented automatically. We compared IA subtypes/volumes across groups, analyzed their predictors, and explored cognitive associations with multivariate regressions. A deep-learning-based IA segmentation tool was also developed and externally validated on 30 new patients. Results IA prevalence did not differ between MS and controls (81.9% vs 74.7%). MS patients showed a shift toward a short IA subtype and reduced IA volume (mean [SD], 146.8 [117.9] vs 230.2 [138.2] mm(3); p<0.0001), worsening across phenotypes. Reduced IA volume was independently associated with medial and posterior thalamic nuclei volumes, but not with white matter lesion load or global atrophy. Among cognitive domains, smaller IA volume was independently associated only with executive dysfunction (OR = 0.89 [0.77-0.99], p = 0.021). The automated IA segmentation tool showed good agreement with manual delineation (median Dice coefficient 0.64; volume correlation r = 0.89). Conclusion IA volume reduction in MS reflects vulnerability of adjacent thalamic nuclei and is associated with executive dysfunction, supporting IA as a marker of thalamic neurodegeneration. The automated IA segmentation tool will enable future larger scale and longitudinal investigations., This work received financial support from the French government in the framework of the University of Bordeaux's France 2030 program/RRI "IMPACT". The work was also conducted in the framework of a grant overseen by the French National Research Agency (ANR) as part of the "France 2030" Program ANR-23-IAHU-0001. The MICROSEP study received financial support from the TRAIL (Translational Research and Advanced Imaging Laboratory) laboratory of excellence (ANR-10-LABX-57). The AUBACOG study was supported by a grant from Genzyme. The PROCOG study was supported by a grant from Roche. This work has been also partially funded, thanks to the project PID2023-152127OB-I00 of the Ministerio de Ciencia e Innovacion of Spain.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Robust deep MRI contrast synthesis using a prior-based and task-oriented 3D network

  • Morell-Ortega, Sergio|||0009-0001-7639-2915
  • Vivó, Roberto|||0000-0002-0751-4114
  • Rubio Navarro, Gregorio
  • DE LA IGLESIA VAYÁ, MARIA DE LOS DESAMPARADOS
  • Manjón Herrera, José Vicente|||0000-0001-6640-927X
  • Ruiz-Perez, Marina
  • Gadea, Marien
  • Aparici, Fernando
  • Thomas Tourdias
  • Mansencal, Boris
  • Coupé, Pierrick
[EN] Magnetic resonance imaging (MRI) is one of the most widely used tools for clinical diagnosis. Depending on the acquisition parameters, different image contrasts can be obtained, providing complementary information about the patient's anatomy and potential pathological findings. However, multiplying such acquisitions requires more time, additional resources, and increases patient discomfort. Consequently, not all image modalities are typically acquired. One solution to obtain the missing modalities is to use contrast synthesis methods. Most existing synthesis methods work with 2D slices due to memory limitations, which produces inconsistencies and artifacts when reconstructing the 3D volume. In this work, we present a 3D deep learning-based approach for synthesizing T2-weighted MR volumes from T1-weighted ones. To preserve anatomical details and enhance image quality, we propose a segmentation-oriented loss function combined with a frequency space information loss. To make the proposed method more robust and applicable to a wider range of image scenarios, we also incorporate a priori information in the form of a multi-atlas. Additionally, we employ a semi-supervised learning framework that improves the model's generalizability across diverse datasets, potentially improving its performance in clinical settings with varying patient demographics and imaging protocols. By integrating prior anatomical knowledge with frequency domain and segmentation loss functions, our approach outperforms state-of-the-art methods, particularly in segmentation tasks. The method demonstrates significant improvements, especially in challenging cases, compared with state-of-the-art approaches., This work has been developed thanks to the project PID2023-152127OB-I00 of the Ministerio de Ciencia, Innovacion e Universidades of Spain. This work benefited from the support of the project HoliBrain of the French National Research Agency (ANR-23-CE45-0020-01). Moreover, this project is supported by the Precision and global vascular brain health institute funded by the France 2030 investment plan as part of the IHU3 initiative (ANR-23-IAHU- 0001). Finally, this study received financial support from the French government in the framework of the University of Bordeaux's France 2030 program/RRI IMPACT and the PEPR StratifyAging. We thank the support of ITACA (Institute of Information and Communication Technologies) at UPV (Universitat Politecnica de Valencia).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Lifespan Tree of Brain Anatomy: Diagnostic Values for Motor and Cognitive Neurodegenerative Diseases

  • Coupé, Pierrick
  • Mansencal, Boris
  • Péran, Patrice
  • Meissner, Wassilios G.
  • Tourdias, Thomas
  • Planche, Vincent
  • Manjón Herrera, José Vicente|||0000-0001-6640-927X
[EN] The differential diagnosis of neurodegenerative diseases, characterized by overlapping symptoms, may be challenging. Brain imaging coupled with artificial intelligence has been previously proposed for diagnostic support, but most of these methods have been trained to discriminate only isolated diseases from controls. Here, we develop a novel machine learning framework, named lifespan tree of brain anatomy, dedicated to the differential diagnosis between multiple diseases simultaneously. It integrates the modeling of volume changes for 124 brain structures during the lifespan with nonlinear dimensionality reduction and synthetic sampling techniques to create easily interpretable representations of brain anatomy over the course of disease progression. As clinically relevant proof-of-concept applications, we constructed a cognitive lifespan tree of brain anatomy for the differential diagnosis of six causes of neurodegenerative dementia and a motor lifespan tree of brain anatomy for the differential diagnosis of four causes of parkinsonism using 37,594 MRIs as a training dataset. This original approach significantly enhanced the efficiency of differential diagnosis in the external validation cohort of 1754 cases, outperforming existing state-of-the-art machine learning techniques. Lifespan tree holds promise as a valuable tool for differential diagnosis in relevant clinical conditions, especially for diseases still lacking effective biological markers., This workwas supported by the French National Research Agency (HoliBrain project, ANR-23-CE45-0020-01; IHU3 initiative, ANR-23-IAHU-0001), the Precision and Global Vascular Brain Health Institute funded by the France 2030 investment plan, and by the French government through the University of Bordeaux's France 2030 program/RRI "IMPACT" and the PEPR StratifyAging and Prodrom-ND. Computational resources were provided by GENCI at IDRIS (allocation 2022-AD011013848R1). Additional support was received from the Ministerio de Ciencia, Innovacion y Universidades de Espana (PID2023-152127OB-I00).