Pasar al contenido principal

EVALUACION MULTIPARAMETRICA DE LA PROGRESION DE LOS TUMORES CEREBRALES BASADA EN LA IA

PID2021-127110OA-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 2021
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): 11
Encontrada(s) 1 página(s)

Dipòsit Digital de Documents de la UAB
Publicación Artículo científico (JournalArticle)

Unlocking Bevacizumab's Potential, rCBVmax as a Predictive Biomarker for Enhanced Survival in Glioblastoma IDH-Wildtype Patients

  • Álvarez-Torres, María del Mar
  • Balañá, Carmen|||0000-0003-0771-0390
  • Fuster-Garcia, Elies|||0000-0002-0716-8960
  • Puig, Josep|||0000-0003-2791-6599
  • García Gómez, Juan Miguel|||0000-0002-3851-1557
Background: Aberrant vascular architecture and angiogenesis are hallmarks of glioblastoma IDH-wildtype, suggesting that these tumors are suitable for antiangiogenic therapy. Bevacizumab was FDA-approved in 2009 following promising results in two clinical trials. However, its use for recurrent glioblastomas remains a subject of debate, as it does not universally improve patient survival. Purposes: In this study, we aimed to analyze the influence of tumor vascularity on the benefit provided by BVZ and propose preoperative rCBVmax at the high angiogenic tumor habitat as a predictive biomarker to select patients who can benefit the most. Methods: Clinical and MRI data from 106 patients with glioblastoma IDH-wildtype have been analyzed. Thirty-nine of them received BVZ, and the remaining sixty-seven did not receive a second-line treatment. The ONCOhabitats method was used to automatically calculate rCBV. Results: We found a median survival from progression of 305 days longer for patients with moderate vascular tumors who received BVZ than those who did not receive any second-line treatment. This contrasts with patients with high-vascular tumors who only presented a median survival of 173 days longer when receiving BVZ. Furthermore, better responses to BVZ were found for the moderate-vascular group with a higher proportion of patients alive at 6, 12, 18, and 24 months after progression. Conclusions: We propose rCBVmax as a potential biomarker to select patients who can benefit more from BVZ after tumor progression. In addition, we propose a threshold of 7.5 to stratify patients into moderate- and high-vascular groups to select the optimal second-line treatment.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Abnormal vascular structure and function within brain metastases is linked to pembrolizumab resistance

  • Kim, Albert E.
  • Lou, Kevin W.
  • Giobbie-Hurder, Anita
  • Chang, Ken
  • Gidwani, Mishka
  • Hoebel, Katharina
  • Patel, Jay B.
  • Cleveland, Mason C.
  • Singh, Praveer
  • Bridge, Christopher P.
  • Rakin Ahmed, Syed
  • Bearce, Benjamin A.
  • Liu, William
  • Lee, Eudocia Q.
  • Fuster García, Elíes|||0000-0002-0716-8960
[EN] Background. We recently conducted a phase 2 trial (NCT028865685) evaluating intracranial efficacy of pembrolizumab for brain metastases (BM) of diverse histologies. Our study met its primary efficacy endpoint and illustrates that pembrolizumab exerts promising activity in a select group of patients with BM. Given the importance of aberrant vasculature in mediating immunosuppression, we explored the relationship between immune checkpoint inhibitor (ICI) efficacy and vascular architecture in the hopes of identifying potential mechanisms of intracranial ICI response or resistance for BM. Methods. Using Vessel Architectural Imaging, a histologically validated quantitative metric for in vivo tumor vascular physiology, we analyzed dual-echo DSC/DCE MRI for 44 patients on trial. Tumor and peri-tumor cerebral blood volume/flow, vessel size, arterial and venous dominance, and vascular permeability were measured before and after treatment with pembrolizumab. Results. BM that progressed on ICI were characterized by a highly aberrant vasculature dominated by largecaliber vessels. In contrast, ICI-responsive BM possessed a more structurally balanced vasculature consisting of both small and large vessels, and there was a trend toward a decrease in under-perfused tissue, suggesting a reversal of the negative effects of hypoxia. In the peri-tumor region, the development of smaller blood vessels, consistent with neo-angiogenesis, was associated with tumor growth before radiographic evidence of contrast enhancement on anatomical MRI. Conclusions. This study, one of the largest functional imaging studies for BM, suggests that vascular architecture is linked with ICI efficacy. Studies identifying modulators of vascular architecture, and effects on immune activity, are warranted and may inform future combination treatments., A.E.K. is supported by an American Brain Tumor Association Basic Research Fellowship In Honor of Paul Fabbri, the William G. Kaelin, Jr., MD, Physician-Scientist Award of the Damon Runyon Cancer Research Foundation, American Association for Cancer Research Breast Cancer Research Fellowship, and American Society of Clinical Oncology Young Investigator Award. E.F.G. was supported by the European Union s Horizon 2020 Programme: Marie Sklodowska-Curie grant agreement No 844646-GLIOHAB and Spanish State Research Agency, Subprogram for Knowledge Generation (PROGRESS, No PID2021-127110OA-I00). P.K.B is supported by the NCI (1R01CA227156-01, 5R21CA220253-02, and 1R01CA244975-01), Damon Runyon Cancer Research Foundation, and Breast Cancer Research Foundation. Y.Y. is supported by the NIH (1R21AG067562-01 and 1R21GM137227-01). K.E.E. is supported by the European Union's Horizon 2020 Programme: ERC Grant Agreement No. 758657-ImPRESS, South-Eastern Norway Regional Health Authority (Grant Agreements No. 2016102, 2017073, 2013069), and the Norwegian Cancer Society and the FRIPRO Grant Agreements No. 261984, 303249). E.R.G. is supported by R01CA211238-01 and R01CA244975.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Treatment-aware Diffusion Probabilistic Model for Longitudinal MRI Generation and Diffuse Glioma Growth Prediction

  • Liu, Qinghui
  • Hovden, Ivar T
  • MacIntosh, Bradley J
  • Grødem, Edvard O.S.
  • Brandal, Peter
  • Sederevicius, Donatas
  • Skogen, Karoline
  • Schellhorn, Till
  • Bjornerud, Atle
  • Emblem, Kyrre E.
  • Fuster García, Elíes|||0000-0002-0716-8960
  • López-Mateu, Carlos
[EN] Diffuse gliomas are malignant brain tumors that grow widespread through the brain. The complex interactions between neoplastic cells and normal tissue, as well as the treatment-induced changes often encountered, make glioma tumor growth modeling challenging. In this paper, we present a novel end-to-end network capable of future predictions of tumor masks and multi-parametric magnetic resonance images (MRI) of how the tumor will look at any future time points for different treatment plans. Our approach is based on cutting-edge diffusion probabilistic models and deep-segmentation neural networks. We included sequential multi-parametric MRI and treatment information as conditioning inputs to guide the generative diffusion process as well as a joint segmentation process. This allows for tumor growth estimates and realistic MRI generation at any given treatment and time point. We trained the model using real-world postoperative longitudinal MRI data with glioma tumor growth trajectories represented as tumor segmentation maps over time. The model demonstrates promising performance across various tasks, including generating high-quality multi-parametric MRI with tumor masks, performing time-series tumor segmentations, and providing uncertainty estimates. Combined with the treatment-aware generated MRI, the tumor growth., This work was supported by the European Union s Horizon 2020 Programme: ERC (European Research Council) Grant Agreement No. 758657-ImPRESS, Helse Sør-Øst Regional Health Authority grants 2021057, 2017073, the Research Council of Norway grants 325971, 261984, and grant PID2021-127110OA-I00 funded by MCIN/AEI/10.13039/501100011033 from Agencia de Investigación de Espana and by ERDF (European Regional Development Fund) A way of making Europe.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Vascular differences between IDH-wildtype glioblastoma and astrocytoma IDH-mutant grade 4 at imaging and transcriptomic levels

  • Álvarez-Torres, María del Mar
  • López-Cerdán, Adolfo
  • Andreu, Zoraida
  • DE LA IGLESIA VAYÁ, MARIA DE LOS DESAMPARADOS
  • Fuster García, Elíes|||0000-0002-0716-8960
  • García García, Francisco|||0000-0003-1488-2764
  • Garcia-Gomez, Juan M
[EN] Global agreement in central nervous system (CNS) tumor classification is essential for predicting patient prognosis and determining the correct course of treatment, as well as for stratifying patients for clinical trials at international level. The last update by the World Health Organization of CNS tumor classification and grading in 2021 considered, for the first time, IDH-wildtype glioblastoma and astrocytoma IDH-mutant grade 4 as different tumors. Mutations in the genes isocitrate dehydrogenase (IDH) 1 and 2 occur early and, importantly, contribute to gliomagenesis. IDH mutation produces a metabolic reprogramming of tumor cells, thus affecting the processes of hypoxia and vascularity, resulting in a clear advantage for those patients who present with IDH-mutated astrocytomas. Despite the clinical relevance of IDH mutation, current protocols do not include full sequencing for every patient. Alternative biomarkers could be useful and complementary to obtain a more reliable classification. In this sense, magnetic resonance imaging (MRI)-perfusion biomarkers, such as relative cerebral blood volume and flow, could be useful from the moment of presurgery, without incurring additional financial costs or requiring extra effort. The main purpose of this work is to analyze the vascular and hemodynamic differences between IDH-wildtype glioblastoma and IDH-mutant astrocytoma. To achieve this, we evaluate and validate the association between dynamic susceptibility contrast-MRI perfusion biomarkers and IDH mutation status. In addition, to gain a deeper understanding of the vascular differences in astrocytomas depending on the IDH mutation, we analyze the transcriptomic bases of the vascular differences., This study was funded by: Grant PID2019-104978RB-I00/AEI/10.13039/501100011033 (ALBATROSS) funded by MCIN/AEI/ 10.13039/501100011033, Grant PID2021-127110OA-I00 (PROGRESS) funded by MCIN/AEI/10.13039/501100011033 (both from Agencia de Investigacion de Espana) and by ERDF A way of making Europe.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Unlocking Bevacizumab&apos, s Potential: rCBVmax as a Predictive Biomarker for Enhanced Survival in Glioblastoma IDH-Wildtype Patients

  • Alvarez-Torres, Maria del Mar
  • Balaña, Carmen
  • Puig, Josep
  • Fuster García, Elíes|||0000-0002-0716-8960
  • Garcia-Gomez, Juan M
[EN] Background: Aberrant vascular architecture and angiogenesis are hallmarks of glioblastoma IDH-wildtype, suggesting that these tumors are suitable for antiangiogenic therapy. Bevacizumab was FDA-approved in 2009 following promising results in two clinical trials. However, its use for recurrent glioblastomas remains a subject of debate, as it does not universally improve patient survival. Purposes: In this study, we aimed to analyze the influence of tumor vascularity on the benefit provided by BVZ and propose preoperative rCBVmax at the high angiogenic tumor habitat as a predictive biomarker to select patients who can benefit the most. Methods: Clinical and MRI data from 106 patients with glioblastoma IDH-wildtype have been analyzed. Thirty-nine of them received BVZ, and the remaining sixty-seven did not receive a second-line treatment. The ONCOhabitats method was used to automatically calculate rCBV. Results: We found a median survival from progression of 305 days longer for patients with moderate vascular tumors who received BVZ than those who did not receive any second-line treatment. This contrasts with patients with high-vascular tumors who only presented a median survival of 173 days longer when receiving BVZ. Furthermore, better responses to BVZ were found for the moderate-vascular group with a higher proportion of patients alive at 6, 12, 18, and 24 months after progression. Conclusions: We propose rCBVmax as a potential biomarker to select patients who can benefit more from BVZ after tumor progression. In addition, we propose a threshold of 7.5 to stratify patients into moderate- and high-vascular groups to select the optimal second-line treatment., M.Á-T was supported by DPI2016-80054-R (Programa Estatal de Promoción del Talento y su Empleabilidad en I + D + i). This work was partially supported by the ALBATROSS project (National Plan for Scientific and Technical Research and Innovation 2017 2020, No. PID2019-104978RB-I00). This study was partially funded by the Fundació La Marató TV3 (665/C/2013). Grant PID2021- 127110OA-I00 (PROGRESS) was funded by MCIN/AEI/10.13039/501100011033 and ERDF, a way of making Europe.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Decreased tissue stiffness in glioblastoma by MR elastography is associated with increased cerebral blood flow

  • Svensson, Siri Fløgstad
  • Latysheva, Anna
  • Fraser-Green, Jorunn
  • Nordhoy, Wibeke
  • Darwish, Omar Isam
  • Hovden, Ivar Thokle
  • Holm, Sverre
  • Vik-Mo, Einar O.
  • Sinkus, Ralph
  • Emblem, Kyrre Eeg
  • Fuster García, Elíes|||0000-0002-0716-8960
[EN] Purpose: Understanding how mechanical properties relate to functional changes in glioblastomas may help explain different treatment response between patients. The aim of this study was to map differences in biome-chanical and functional properties between tumor and healthy tissue, to assess any relationship between them and to study their spatial distribution. Methods: Ten patients with glioblastoma and 17 healthy subjects were scanned using MR Elastography, perfusion and diffusion MRI. Stiffness and viscosity measurements G' and G '', cerebral blood flow (CBF), apparent diffusion coefficient (ADC) and fractional anisotropy (FA) were measured in patients' contrast-enhancing tumor, necrosis, edema, and gray and white matter, and in gray and white matter for healthy subjects. A regression analysis was used to predict CBF as a function of ADC, FA, G' and G ''. Results: Median G' and G' ' in contrast-enhancing tumor were 13% and 37% lower than in normal-appearing white matter (P < 0.01), and 8% and 6% lower in necrosis than in contrast-enhancing tumor, respectively (P < 0.05). Tumors showed both inter-patient and intra-patient heterogeneity. Measurements approached values in normalappearing tissue when moving outward from the tumor core, but abnormal tissue properties were still present in regions of normal-appearing tissue. Using both a linear and a random-forest model, prediction of CBF was improved by adding MRE measurements to the model (P < 0.01). Conclusions: The inclusion of MRE measurements in statistical models helped predict perfusion, with stiffer tissue associated with lower perfusion values., Grant Support: We gratefully acknowledge support from the European Union's Horizon 2020 Programme: ERC Grant Agreement No. 758657-ImPRESS, Research and Innovation Grant Agreements No. 668039-FORCE and Marie Sklodowska-Curie grant agreement (No 844646-GLIOHAB), South-Eastern Norway Regional Health Authority (Grant Agreement No. 2017073, 2013069 and 2021057), The Research Council of Norway FRIPRO (Grant Agreement No. 261984), the National Institutes of Health R21 grant (Grant Agreement No. EB030757), and the German Research Foundation (DFG, SCHR 1542/1-1).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

The prognostic relevance of a gene expression signature in MRI-defined highly vascularized glioblastoma

  • Montosa-Micó, Víctor
  • Gil-Terrón-Rodríguez, Francisco Javier|||0000-0002-3536-6462
  • Gómez-Mahiques, María
  • López-Mateu, Carlos
  • Garcia-Gomez, Juan M
  • Fuster García, Elíes|||0000-0002-0716-8960
  • Álvarez-Torres, María del Mar
  • Burgos-Panadero, Rebeca
[EN] Background The vascular heterogeneity of glioblastomas (GB) remains an important area of research, since tumor progression and patient prognosis are closely tied to this feature. With this study, we aim to identify gene expression profiles associated with MRI-defined tumor vascularity and to investigate its relationship with patient prognosis. Methods The study employed MRI parameters calculated with DSC Perfusion Quantification of ONCOhabitats glioma analysis software and RNA-seq data from the TCGA-GBM project dataset. In our study, we had a total of 147 RNA-seq samples, which 15 of them also had MRI parameter information. We analyzed the gene expression profiles associated with MRI-defined tumor vascularity using differential gene expression analysis and performed Log-rank tests to assess the correlation between the identified genes and patient prognosis. Results The findings of our research reveal a set of 21 overexpressed genes associated with the high vascularity pattern. Notably, several of these overexpressed genes have been previously implicated in worse prognosis based on existing literature. Our log-rank test further validates that the collective upregulation of these genes is indeed correlated with an unfavorable prognosis. This set of genes includes a variety of molecules, such as cytokines, receptors, ligands, and other molecules with diverse functions. Conclusions Our findings suggest that the set of 21 overexpressed genes in the High Vascularity group could potentially serve as prognostic markers for GB patients. These results highlight the importance of further investigating the relationship between the molecules such as cytokines or receptors underlying the vascularity in GB and its observation through MRI and developing targeted therapies for this aggressive disease., This work was funded by Grant PID2021-127110OA-I00 (PROGRESS) funded by MCIN/AEI/10.13039/501100011033 and by ERDF, Beca Margarita Salas funded by Ministerio de Ciencia, Innovacion y Universidades. Grant ID: MS21-046, SINUE (INNEST/2022/87 Agencia Valenciana de la Innovacion), ALBATROSS (PID2019-104978RB-I00), Programa INVESTIGO (INVEST/2022/298) and Grant CIAICO/2022/064 (Lalaby-Glio) funded by Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Interpreting deep learning models for glioma survival classification using visualization and textual explanations.

  • Osadebey, Michael
  • Liu, Qinghui
  • Emblem, Kyrre E.
  • Fuster García, Elíes|||0000-0002-0716-8960
[EN] Background Saliency-based algorithms are able to explain the relationship between input image pixels and deeplearning model predictions. However, it may be difcult to assess the clinical value of the most important image features and the model predictions derived from the raw saliency map. This study proposes to enhance the interpret¿ ability of saliency-based deep learning model for survival classifcation of patients with gliomas, by extracting domain knowledge-based information from the raw saliency maps. Materials and methods Our study includes presurgical T1-weighted (pre- and post-contrast), T2-weighted and T2-FLAIR MRIs of 147 glioma patients from the BraTs 2020 challenge dataset aligned to the SRI 24 anatomical atlas. Each image exam includes a segmentation mask and the information of overall survival (OS) from time of diag¿ nosis (in days). This dataset was divided into training (n = 118) and validation (n = 29) datasets. The extent of surgi¿ cal resection for all patients was gross total resection. We categorized the data into 42 short (mean µ = 157 days), 30 medium (µ = 369 days), and 46 long (µ = 761 days) survivors. A 3D convolutional neural network (CNN) trained on brain tumour MRI volumes classifed all patients based on expected prognosis of either short-term, mediumterm, or long-term survival. We extend the popular 2D Gradient-weighted Class Activation Mapping (Grad-CAM), for the generation of saliency map, to 3D and combined it with the anatomical atlas, to extract brain regions, brain volume and probability map that reveal domain knowledge-based information. Results For each OS class, a larger tumor volume was associated with a shorter OS. There were 10, 7 and 27 tumor locations in brain regions that uniquely associate with the short-term, medium-term, and long-term survival, respec¿ tively. Tumors located in the transverse temporal gyrus, fusiform, and palladium are associated with short, medium and long-term survival, respectively. The visual and textual information displayed during OS prediction highlights tumor location and the contribution of diferent brain regions to the prediction of OS. This algorithm design feature assists the physician in analyzing and understanding diferent model prediction stages. Conclusions Domain knowledge-based information extracted from the saliency map can enhance the interpret¿ ability of deep learning models. Our fndings show that tumors overlapping eloquent brain regions are associated with short patient survival., This work was supported by The European Union s Horizon 2020 Programmes ERC Grant Agreement No. 758657-ImPRESS, and Marie Sk&#322;odowska-Curie grant agreement (No 844646-GLIOHAB). The Norwegian Cancer Society and the Research Council of Norway Grants 303249, 261984, 325971. South-Eastern Norway Regional Health Authority Grants 2021057, 2013069, 2017073. Elies Fuster-Garcia was supported by Spanish State Research Agency, Subprogram for Knowledge Generation (PROGRESS, No PID2021-127110OA-I00).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Automated delineation of putative non-contrast-enhancing tumor in glioblastoma: Prognostic insights

  • Gómez-Mahiques, María
  • Garcia-Gomez, Juan M
  • Montosa-Micó, Víctor
  • López-Mateu, Carlos
  • Gil-Terrón-Rodríguez, Francisco Javier|||0000-0002-3536-6462
  • Fuster García, Elíes|||0000-0002-0716-8960
  • Svensson, SF
  • Mireles, Eduardo Erasmo Mendoza
  • Vik-Mo, Einar Osland
  • Emblem, Kyrre E.
  • Balana-Quintero, Carme
  • Puig, Josep
  • Alenda, Cristina
  • Martinez-Saez, Elena
  • Martinez-Ricarte, Fran
  • Quiros-Marti, Marta
  • Quilis-Quesada, Vicent
[EN] Background Precise delineation of non-contrast-enhancing tumor (nCET) in glioblastoma (GB) is critical for maximal safe resection, yet routine imaging cannot reliably separate infiltrative tumor from vasogenic edema. The aim of this study was to develop and validate an automated method to identify peritumoral subregions compatible with nCET and assess its prognostic value.Methods Pre-operative T2-weighted and FLAIR MRI from 940 patients with newly diagnosed GB in four multicenter cohorts were analyzed. A deep-learning model segmented enhancing tumor, edema and necrosis; a non-local, spatially varying finite mixture model was applied to identify edema subregions characterized by relatively lower FLAIR hyperintensity, hypothesized to reflect nCET-related tissue. The ratio of these subregions to total edema volume defined the T2/FLAIR Heterogeneity Index (TFHI). Associations between TFHI and overall survival (OS) were examined with Kaplan-Meier curves and multivariable Cox regression.Results Higher TFHI values stratified patients with shorter OS. In the NCT03439332, TFHI above the optimal threshold was associated with a twofold increased hazard of death (hazard ratio [HR] 2.07, 95 % confidence interval 1.33-3.21; P = .0013) and a reduction in median survival of 98 days. Significant, though smaller, prognostic effects were confirmed in GLIOCAT & BraTS (HR= 1.37; P = .047), OUS (HR = 1.37; P = .0032) and pooled analysis (HR= 1.26; P = .0008). TFHI remained an independent predictor after adjustment for age, extent of resection and MGMT methylation.Conclusions We present a reproducible, server-hosted tool for automated identification of imaging-defined, putative nCET-related peritumoral subregions and TFHI biomarker extraction that enables independent prognostic stratification. This approach provides a quantitative framework for studying peritumoral heterogeneity in GB., Spanish Ministry of Science, Innovation and Universities (FPU23/02431); European Research Council (758657-ImPRESS); Helse Sor-& Oslash;st Regional Health Authority (2021057, 2017073); Research Council of Norway (325971, 261984, 303249, 32543); Fundacio La Marato de TV3 (665/C/2013); Instituto de Salud Carlos III (PI18/01062); Generalitat Valenciana (INVEST/2022/298, CIAICO/2022/064); Agencia Valenciana de la Innovacion (INNEST/2022/87); Universitat Politecnica de Valencia (PAID-01-24); This work was funded by Grant PID2021-127110OA-I00 (PROGRESS) funded by MCIN/AEI/10.13039/501100011033 and by ERDF. In addition, Project ECHIDNA (PID2024-156848OB-I00) and LAB2OR (PDC2025-165365-I00) funded by MICIU/AEI/10.13039/501100011033 and ERDF/EU. Funding for open access charge: CRUE-Universitat Politecnica de Valencia.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Exploring the Trade-Off between generalist and specialized Models: A center-based comparative analysis for glioblastoma segmentation

  • Gil-Terrón-Rodríguez, Francisco Javier|||0000-0002-3536-6462
  • Ferri-Borredà, Pablo|||0000-0002-3083-1169
  • Montosa-Micó, Víctor
  • Gómez-Mahiques, María
  • López-Mateu, Carlos
  • Garcia-Gomez, Juan M
  • Fuster García, Elíes|||0000-0002-0716-8960
  • Martí, Pau
[EN] Introduction Inherent variations between inter-center data can undermine the robustness of segmentation models when applied at a specific center (dataset shift). We investigated whether specialized center-specific models are more effective compared to generalist models based on multi-center data, and how center-specific data could enhance the performance of generalist models within a particular center using a fine-tuning transfer learning approach. For this purpose, we studied the dataset shift at center level and conducted a comparative analysis to assess the impact of data source on glioblastoma segmentation models. Methods & Materials The three key components of dataset shift were studied: prior probability shift¿variations in tumor size or tissue distribution among centers; covariate shift¿inter-center MRI alterations; and concept shift¿different criteria for tumor segmentation. BraTS 2021 dataset was used, which includes 1251 cases from 23 centers. Thereafter, 155 deep-learning models were developed and compared, including 1) generalist models trained with multi-center data, 2) specialized models using only center-specific data, and 3) fine-tuned generalist models using center-specific data. Results The three key components of dataset shift were characterized. The amount of covariate shift was substantial, indicating large variations in MR imaging between different centers. Glioblastoma segmentation models tend to perform best when using data from the application center. Generalist models, trained with over 700 samples, achieved a median Dice score of 88.98%. Specialized models surpassed this with 200 cases, while fine-tuned models outperformed with 50 cases. Conclusions The influence of dataset shift on model performance is evident. Fine-tuned and specialized models, utilizing data from the evaluated center, outperform generalist models, which rely on data from other centers. These approaches could encourage medical centers to develop customized models for their local use, enhancing the accuracy and reliability of glioblastoma segmentation in a context where dataset shift is inevitable., This work was funded by Grant PID2021-127110OA-I00 (PROGRESS) funded by MCIN/AEI/ 10.13039/501100011033 and by ERDF, SINUE (INNEST/2022/87 Agencia Valenciana de la Innovacion, Spain), ALBATROSS (PID2019-104978RB-I00) , Programa INVESTIGO (INVEST/2022/298) and Grant CIAICO/2022/064 (Lalaby-Glio) funded by Generalitat Valenciana, Spain.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Biomechanical mapping of tumor growth: A novel method to quantify glioma infiltration and mass effect

  • López-Mateu, Carlos
  • Gómez-Mahiques, María
  • Montosa-Micó, Víctor
  • Gil-Terrón-Rodríguez, Francisco Javier|||0000-0002-3536-6462
  • Garcia-Gomez, Juan M
  • Fuster García, Elíes|||0000-0002-0716-8960
  • Sederevicius, Donatas
  • Emblem, Kyrre E.
[EN] Background: Glioblastoma (GBM) growth can alter surrounding brain tissue through location-dependent physiological changes. Two main growth phenotypes¿(I) infiltrative, characterized by diffuse invasion with minimal mass effect, and (II) proliferative, characterized by pronounced tissue compression¿are recognized, but their quantitative characterization and prognostic impact remain poorly explored. Purpose: To develop and validate a novel MRI-based biomarker, the Dynamic Infiltration Rate (DIR), that quantitatively assesses the balance between tumor volume expansion and peritumoral compression, and to evaluate its prognostic ability for stratifying patients based on overall survival (OS). Methods: The DIR was defined as the ratio between tumor-volume enlargement and mass-effect-induced peritumoral compression. Technical validation was conducted using synthetic datasets with known ground truth spanning realistic infiltrative-proliferative spectra. Clinically, patients were dichotomized into high- and low-infiltration groups using a threshold optimized by maximizing the log-rank statistic for OS. Prognostic evaluation included multivariate Cox regression adjusted for age, sex, and MGMT methylation status. Results: The synthetic dataset validation demonstrated high concordance with ground truth (R² = 0.89). Clinical evaluation indicated significantly improved OS in the low DIR group (median = 35.2 weeks) compared to the high DIR group (median = 16.0 weeks; p = 0.0001). DIR effectively stratified patients based on survival (log-rank p < 0.001, HR = 2.49) and remained an independent prognostic factor on multivariate analysis (HR = 1.45, 95% CI 1.07¿1.85; p = 0.0159). Conclusions: The DIR is a novel and robust quantitative MRI biomarker capable of distinguishing between proliferative and infiltrative GBM phenotypes, independently predicting OS. Early phenotype identification could facilitate personalized treatment strategies and individualized follow-up scheduling., MCIN/AEI (Ministerio de Ciencia, Innovación y Universidades / Agencia Estatal de Investigación), Grant/Award Numbers: PID2024156848OB-I00, PID2021-127110OA-I00, PDC2025-165365-I00, FPU23/02431; Generalitat Valenciana, Grant/Award Numbers: INVEST/2022/298 (ProgramaINVESTIGO), CIAICO/2022/064(Lalaby-Glio), INNEST/2022/87(SINUÉ); European Research Council, Grant/Award Numbers: 758657-ImPRESS, 101213953-CHRONOS; Helse Sør-Øst RHF, Grant/Award Numbers: 2021057, 2017073, 2023050; Research Council of Norway, Grant/Award Numbers: 325971, 261984, 303249, 32543, 356066; Norwegian Cancer Society, Grant/Award Number: 309012 - SAFARI; Universitat Politècnica de València, Grant/Award Number: PAID-01-24(Subprogram1)