Scientific Publications
Learn more about our team's latest work in neuroimaging, medical imaging, and clinical trials.
2023
Towards an enhanced characterization of white matter pathologies with Q-FiberMapper 2.0
Tommy Boshkovski, Óscar Peña-Nogales, Paulo Rodrigues, Vesna Prčkovska, Kire Trivodaliev
FiberMapper 2.0 facilitates automatic (pre-) processing of MRI data, reconstructs major white matter tracts and derives imaging biomarkers along the tracts sensitive to a wide range of microstructural properties. Additionally, Q-FiberMapper 2.0 implements tract lesion burden analysis which together with the quantitative MR imaging biomarkers could be very useful for early diagnosis, evaluation and monitoring of different neurodegenerative and neuroinflammatory diseases that affect the white matter such as traumatic brain injury, multiple sclerosis, and others.
Incorporation of the central vein sign into the International Panel criteria
Moein Amin, ..., Paulo Rodrigues, ... Vesna Prchkovska, ..., Marc Ramos,... Daniel Ontaneda (2023)
The diagnosis of multiple sclerosis (MS) relies on establishing dissemination in time (DIT) and dissemination in space (DIS) as codified in the 2017 International Panel criteria (IP2017). Although sensitive, the IP2017 has a misdiagnosis rate of 20%, primarily attributed to incorrect application of MRI criteria. The central vein sign (CVS) is a putative diagnostic biomarker for MS that may increase specificity and diagnostic accuracy, but how to optimally incorporate CVS with the IP2017 criteria is not established.
Assure Clinical Trials Quality while ensuring efficiency: Automatic Data Classification and Quality Protocol Adherence of Medical Imaging Data
Óscar Peña-Nogales, Evie Neylon, Marc Ramos, Tommy Boshkovsky, paulo Rodrigues, Vesna Prchkovska, Kire Trivodaliev
Imaging biomarkers can be derived from multiple medical imaging modalities and are helpful for assessing the safety and effectiveness of clinical trials. However, the diversity of medical image modalities creates complexity for archival systems. Furthermore, acquisitions must be in compliance with the imaging charter to ensure the trial’s quality. Thus, we propose an automatic data classification and quality protocol adherence (QPA) approach for medical imaging data named Smart-Uploader.
A systematic review of (semi-)automatic quality control of T1-weighted MRI scans
Janine Hendriks, Henk-Jan Mutsaerts, Richard Joules, Óscar Peña-Nogales, Paulo R. Rodrigues, Robin Wolz, George L. Burchell, Frederik Barkhof & Anouk Schrantee
Artifacts in magnetic resonance imaging (MRI) scans degrade image quality and thus negatively affect the outcome measures of clinical and research scanning. Considering the time-consuming and subjective nature of visual quality control (QC), multiple (semi-)automatic QC algorithms have been developed. This systematic review presents an overview of the available (semi-)automatic QC algorithms and software packages designed for raw, structural T1-weighted (T1w) MRI datasets. The objective of this review was to identify the differences among these algorithms in terms of their features of interest, performance, and benchmarks.
2022
Machine Learning-supported MRI Radiomics to predict the Volumetric Response in Meningiomas after Gamma Knife Radiosurgery
H. Speckter, …, K. Trivodaliev, …, P. Stoeter. (2022). MRI radiomics in the prediction of the volumetric response in meningiomas after gamma knife radiosurgery. Journal of Neuro-Oncology, June.
This is the first report of a strong association between MRI radiomic features and volumetric meningioma response to radiosurgery. The clinical importance of the early and reliable prediction of meningioma responsiveness to radiosurgery is based on its potential to aid individualized therapy decision making.
Insights from the IronTract challenge: optimal methods for mapping brain pathways from multi-shell diffusion MRI
C. Maffeia, …T. Boshkovski, …, M. Ramos, P. Rodrigues, V. Prčkovska, Robert, A. Yendikia. (2022). Insights from the IronTract challenge: Optimal methods for mapping brain pathways from multi-shell diffusion MRI. NeuroImage, August.
Limitations in the accuracy of brain pathways reconstructed by diffusion MRI (dMRI) tractography have received considerable attention. While the technical advances spearheaded by the Human Connectome Project (HCP) led to significant improvements in dMRI data quality, it remains unclear how these data should be analyzed to maximize tractography accuracy. Over a period of two years, we have engaged the dMRI community in the IronTract Challenge, which aims to answer this question by leveraging a unique dataset.
Q-FiberMapper - A framework for tractography and tractometry of clinical data
T. Boshkovski, G. Gallardo, O. Nogales, M. Ramos, K. Trivodaliev, P. Rodrigues, V. Prchkovska. 2022. Q-FiberMapper - A framework for tractography and tractometry of clinical data. In ISMRM 2022.
As the COVID-19 pandemic continues, evidence based clinical guidance for managing the care of people with multiple sclerosis (MS) is an ongoing concern. In recent months, data from cohorts of people with MS has indicated that certain demographic and clinical characteristics, including use of some disease-modifying therapies (DMTs), leads to worse outcomes from SARS-CoV-2 infection. Clinician-reported data from 32 countries were aggregated into a dataset of 5,543 patients who had suspected or confirmed COVID-19. Of 5,543 patients in the clinician-reported dataset, 909 with suspected and 4,634 with confirmed COVID-19 were included in the analysis. Previous demographic findings were confirmed: male sex, older age, progressive MS, and higher disability were associated with worse outcomes from SARS-CoV-2 infection. Use of anti-CD20 DMTs (ocrelizumab and rituximab) was associated with worse COVID-19 outcomes.
Severity of COVID19 infection among patients with multiple sclerosis treated with interferon-β
S. Simpson, …, L. Peeters, …, P. Rodrigues, V. Prčkovska, …, T. Kalincik. (2022). Severity of COVID19 infection among patients with multiple sclerosis treated with interferon-β. Multiple Sclerosis and Related Disorders Elsevier, July.
Interferon-β, a disease-modifying therapy (DMT) for MS, may be associated with less severe COVID-19 in people with MS. Among 5,568 patients (83.4% confirmed COVID-19), interferon-treated patients had lower risk of severe COVID-19 compared to untreated, but not to glatiramer-acetate, dimethyl-fumarate, or pooled other DMTs. In comparison to other DMTs, we did not find evidence of protective effects of interferon-β on the severity of COVID-19, though compared to the untreated, the course of COVID19 was milder among those on interferon-β. This study does not support the use of interferon-β as a treatment to reduce COVID-19 severity in MS.
2021
Updated results of the COVID-19 in MS global data sharing initiative validate consistent associations of anti-CD20 and other reported risk factors with severe COVID-19 outcomes
S. Simpson-Yap, …, P. Rodrigues, …, L. Peeters. (2021). Updated results of the COVID-19 in MS global data sharing initiative validate consistent associations of anti-CD20 and other reported risk factors with severe COVID-19 outcomes. In ECTRIMS 2021 – Late Breaking News Oral Presentations.
As the COVID-19 pandemic continues, evidencebased clinical guidance for managing the care of people with multiple sclerosis (MS) is an ongoing concern. In recent months, data from cohorts of people with MS has indicated that certain demographic and clinical characteristics, including use of some disease-modifying therapies (DMTs), leads to worse outcomes from SARS-CoV-2 infection. The COVID-19 in MS global data sharing initiative, which now includes over 4,500 confirmed COVID19 cases in people with MS, gives the opportunity to corroborate previous findings with greater certainty.
Associations of Disease-Modifying Therapies With COVID-19 Severity in Multiple Sclerosis
Steve Simpson-Yap, ... Nikola Lazovski, ... Liesbet Peeters. 2021. Associations of Disease-Modifying Therapies With COVID-19 Severity in Multiple Sclerosis. Neurology, October.
People with multiple sclerosis (MS) are a vulnerable group for severe COVID- 19, particularly those taking immunosuppressive disease-modifying therapies (DMTs). We examined the characteristics of COVID-19 severity in an international sample of people with MS. Using the largest cohort of people with MS and COVID-19 available, we demonstrated consistent associations of rituximab with increased risk of hospitalization, ICU admission, and need for artificial ventilation and of ocrelizumab with hospitalization and ICU admission.
Central Vein Sign: A Diagnostic Biomarker in Multiple Sclerosis (CAVS-MS) Study Protocol for a Prospective Multicenter Trial
D. Ontaneda, P. Sati,... D. Moreno-Dominguez, M. Ramos,... P. Rodrigues,... N. L. Sicotte. 2021. Central Vein Sign: A Diagnostic Biomarker in Multiple Sclerosis (CAVS-MS) Study Protocol for a Prospective Multicenter Trial. NeuroImage: Clinical 32(January).
The specificity and implementation of current MRI-based diagnostic criteria for multiple sclerosis (MS) are imperfect. Approximately 1 in 5 of individuals diagnosed with MS are eventually determined not to have the disease, with overreliance on MRI findings a major cause of MS misdiagnosis. The central vein sign (CVS), a proposed MRI biomarker for MS lesions, has been extensively studied in numerous cross sectional studies and may increase diagnostic specificity for MS. ...
On the generalizability of diffusion MRI signal representations across acquisition parameters, sequences and tissue types: chronicles of the MEMENTO challenge
Alberto De Luca,... Irina Sanchez, Vesna Prchkovska, Paulo Rodrigues,... Kurt G Schilling. 2021. On the Generalizability of Diffusion MRI Signal Representations across Acquisition Parameters, Sequences and Tissue Types: Chronicles of the MEMENTO Challenge. Preprint. Neuroscience.
Diffusion MRI (dMRI) has become an invaluable tool to assess the microstructural organization of brain tissue. Depending on the specific acquisition settings, the dMRI signal encodes specific properties of the underlying diffusion process. In the last two decades, several signal representations have been proposed to fit the dMRI signal and decode such properties. Most methods, however, ...
New insights from the IronTract challenge: Simple post-processing enhances the accuracy of diffusion tractography
Chiara Maffei,... Santi Puch, Marc Ramos, Nikola Lazovski, Paulo Rodrigues, Vesna Prchkovska,... , Anastasia Yendiki. (2021) New insights from the IronTract challenge: Simple post-processing enhances the accuracy of diffusion tractography. In ISMRM & SMRT Annual Meeting.
We present results from round 2 of IronTract, the first challenge to evaluate the accuracy of tractography using i) tracer injections and diffusion MRI from the same macaque brains, and ii) DSI and HCP two-shell diffusion acquisition schemes. In round 1, only two teams achieved similarly high performance between the two different injection sites that we used for training and validation. Here we investigate the extent to which this was due to the pre- and post-processing used by those teams. We show that, when other teams use the same pre- and post-processing, their accuracy and robustness can improve as well.
Automated workflow for volumetric assessment of signal intensity ratio on T1-weighted MR images after multiple gadolinium administrations
Chia-Ying Liu, Marc Ramos, David Moreno-Dominguez, Vesna Prckovska, Paulo Rodrigues, …, Jacob Agris. (2021) Automated workflow for volumetric assessment of signal intensity ratio on T1-weighted MR images after multiple gadolinium administrations. Journal of Medical Imaging 8(1).
Repeated injections of linear gadolinium-based contrast agent (GBCA) have shown correlations with increased signal intensities (SI) on unenhanced T1-weighted (T1w) images. Assessment is usually performed manually on a single slice and the SI as an average of a freehand region-of-interest is reported. We aim to develop a fully automated software that segments and computes SI ratio of dentate nucleus (DN) to pons (DN/P) and globus pallidus (GP) to thalamus (GP/T) for the assessment of gadolinium presence in the brain after a serial GBCA administrations.
The Central Vein Sign in Multiple Sclerosis (CAVS-MS): A North American Imaging in MS (NAIMS) Cooperative Prospective Diagnostic Biomarker Study
D. Ontaneda, …,D. Moreno, …, P. Rodrigues, …, N. Sicotte (2021) The Central Vein Sign in Multiple Sclerosis (CAVS-MS): A North American Imaging in MS (NAIMS) Cooperative Prospective Diagnostic Biomarker Study. In ACTRIMS 2021 Forum. American Committee for Treatment and Research in Multiple Sclerosis.
The CAVS-MS study intends to definitively establish the CVS as a diagnostic biomarker for MS which can be applied broadly to individuals presenting for evaluation of possible MS. The results of this study will impact diagnostic sensitivity in patients with typical presentations and potentially validate the specificity of CVS in those with atypical presentations, thereby reducing unnecessary costs associated with misdiagnosis.
A Multicenter Evaluation of the Diagnostic Performance of the Central Vein Sign Using Simplified Algorithms
L. Daboul, …, D. Moreno-Dominguez, P. Rodrigues, … , P. Sati, D. Ontaneda (2021) A Multicenter Evaluation of the Diagnostic Performance of the Central Vein Sign Using Simplified Algorithms. In ACTRIMS 2021 Forum. American Committee for Treatment and Research in Multiple Sclerosis.
The objective is to evaluate the sensitivity and specificity of simplified algorithms for assessing the CVS using FLAIR* for MS diagnosis. Images were uploaded to a cloud server (QMENTA). Trained raters selected up to 6 lesions meeting NAIMS criteria on pre- and post-contrast FLAIR* images. The diagnostic performance of the CVS was evaluated at thresholds of 1 CVS+ lesion (Select-1*) up to 6 (Select-6*). The conclusion is that simplified CVS algorithms rated by clinical neurologists can accurately discriminate MS and non-MS cases.
2020
The IronTract challenge: Validation and optimal tractography methods for the HCP diffusion acquisition scheme.
…, Nikola Lazovski, Albert Puente, Matt Rowe, Irina Sanchez, Vesna Prchkovska, …, Anastasia Yendiki. (2020) The IronTract challenge: Validation and optimal tractography methods for the HCP diffusion acquisition scheme. In ISMRM 2020: 28th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
Results from IronTract, the first challenge to evaluate tractography on the two-shell diffusion scheme of the Human Connectome Project (HCP). Accuracy was evaluated by comparison to tracer injections in the same macaque brains as the diffusion data. Training and validation datasets involved different injection sites.
Measures of lesion load on white matter fiber bundles for damage assessment
Garcia, G., Moreno-Dominguez, D., Ramos, M., Rowe, M. (2020) Measures of lesion load on white matter fiber bundles for damage assessment. In ISMRM 2020: 28th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
Measures of lesion load are useful to study how damage in the white matter present in different pathologies relates to changes in cognitive function. So far, most research focus on global or local volumetric metrics, an approach that exhibits limitations in cases where small lesions in specific places cause major damages. In this work, we propose the combined use of a set of metrics that measure different aspects of the lesions over major white matter bundles. We expose how these metrics provide complementary information and discuss how its usefulness could be assessed in future work.
COVID-19 in people with multiple sclerosis: A global data-sharing initiative
Peeters, L., …, Rodrigues, P., Lazovski, N., …, Rijke, N. (2020) COVID-19 in people with multiple sclerosis: A global data-sharing initiative. Multiple Sclerosis Journal 26(10).
High-quality data is needed to assess the determinants for COVID-19 severity in people with MS. Our mission is to scale-up COVID-19 data collection efforts and provide the MS community with data-driven insights as soon as possible. The Multiple Sclerosis International Federation and the Multiple Sclerosis Data Alliance have teamed up with QMENTA and multiple partners to establish a global data-sharing initiative.
CentrAl Vein Sign in the early diagnosis of Multiple Sclerosis (CAVS-MS): a pilot study
Sati, P., Moreno-Dominguez, D., Rodrigues, P., Shinohara, R., Solomon. A., Reich, D.S., Ontaneda, D. (2020) Central Vein Sign in the early diagnosis of Multiple Sclerosis (CAVS-MS): A pilot study. In ACTRIMS 2020 Forum. American Committee for Treatment and Research in Multiple Sclerosis.
The central vein sign in MS (CVS) has been found to be present in a majority of white matter lesions in MS patients, with significantly lower numbers in other conditions with white matter lesions. To demonstrate the feasibility of a prospective multi-center observational study for detection of CVS through the application of FLAIR* images across 10 sites from within the North American Imaging in MS Cooperative (NAIMS).
2019
The structural disconnectome of the MS brain
Prčkovska, V., Rodrigues, P., Ramos, M., Moreno-Dominguez, D., Puch, S., Rowe, M., Andorra, M. & Villoslada, P. (2019). The structural disconnectome of the MS brain. ECTRIMS 2019: 35th Congress of the European Committee for Treatment and Research in Multiple Sclerosis.
Brain damage in MS involves the transection of WM tracts by lesions, creating a disconnection syndrome which defines the clinical phenotype. However, we lack detailed characterization of tract-specific damage in MS, as well as understanding of its contribution to clinical disability. In this work, we retrieved MS dMRI-based structural connectomes and analyzed their connectivity strength differences. We also developed multiple linear regression models that explained a significant proportion of clinical scales.
Few-shot Learning with Deep Triplet Networks for Brain Imaging Modality Recognition
Puch, S., Sánchez, I., Rowe, M. (2019). Few-shot Learning with Deep Triplet Networks for Brain Imaging Modality Recognition”. In Medical Image Learning with Less Labels and Imperfect Data 2019.
This paper details a deep neural network model based on the well-established Triplet Networks to recognize brain imaging modalities based on image intensities in a scenario where some of the classes have a limited number of instances. The paper also shows an evaluation of the proposed model with noisy cases and how it can be used to provide an initial measure of uncertainty to recognize out-of-class samples.
Tractography Reproducibility Challenge with Empirical Data (TraCED): The 2017 ISMRM Diffusion Study Group Challenge
(in press) Nath, V. Schilling, K. G., et int. (…, Rowe, M., Rodrigues, P., Prčkovska, V., …), & Landman, B. A. (2019). “Tractography Reproducibility Challenge with Empirical Data (TraCED): The 2017 ISMRM Diffusion Study Group Challenge”. Journal of Magnetic Resonance Imaging.
This paper details the competition organised by Bennett Landmann and held at the diffusion study group at ISMRM 2017 in which a number of teams competed to produce the most reproducible tractography on a set of specified white matter structures. QMENTA’s entry was the 3rd most reproducible method.
Advanced medical imaging handling and analysis in clinical trials
Rowe, M. (2019). Advanced medical imaging handling and analysis in clinical trials. In Alzheimer’s & Parkinson’s Diseases Congress – AD/PD Lisbon, 2019.
Presentation on the acceleration of handling imaging data in clinical trials through a cloud-based centralized data repository and management system.
Global Planar Convolutions for Improved Context Aggregation in Brain Tumor Segmentation
Puch, S., Sánchez, I., Hernández, A., Piella, G., Prc̆kovska, V. (2019). Global Planar Convolutions for Improved Context Aggregation in Brain Tumor Segmentation. In: Crimi A., Bakas S., Kuijf H., Keyvan F., Reyes M., van Walsum T. (eds) Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. BrainLes 2018. Lecture Notes in Computer Science, vol 11384. Springer, Cham.
In this paper, we introduce the Global Planar Convolution module, a building-block for Convolutional Neural Networks that enhances context perception capabilities, and we apply them to the task of delineating gliomas and their main compartments.
Global Planar Convolutions for improved context aggregation in Brain Tumor Segmentation with MR images
Puch, S., Sánchez, I., Hernández, A., Piella, G., Rodrigues, P., & Prčkovska, V. (2019). Global planar convolutions for improved context aggregation in brain tumor segmentation with MR images. In ISMRM 2019: 27th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
This publication describes our preliminary work on a novel architecture based on Global Planar Convolutions applied to the task of brain tumor segmentation.
Semi-automatic cloud-based workflow for evaluating the central vein sign for MS diagnosis in a multicenter clinical setting
Moreno-Dominguez, D., Ramos, M., Reich, D. S., Ontaneda, D., Rodrigues, P., & Sati, P. (2019). Semi-automatic cloud-based workflow for evaluating the central vein sign for MS diagnosis in a multicenter clinical setting. In ISMRM 2019: 27th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
In this work, we developed a semi-automatic cloud-based workflow for evaluating the clinical value of the central vein sign for MS diagnosis using FLAIR* in a multicenter setting. This novel workflow is a powerful tool that has the potential to significantly accelerate clinical research imaging studies in MS.
Automated cloud-based workflow for quantification of MRI signal intensity – initial real-world clinical validation
Ramos, M., Prčkovska, V., Rodrigues, P., Wang, J., Moser, F., Blank, M., Agarwal, S., Agris, J., & Moreno-Dominguez, D. (2019). Automated cloud-based workflow for quantification of MRI signal intensity – initial real-world clinical validation. In ISMRM 2019: 27th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
We present a fully automatic workflow which accelerates the investigation of contrast agent depositions such as Gadolinium by extracting the T1-weighted modal intensity value and applies appropriate corrections and normalizations to allow comparison across acquisitions and protocols. Automatic results matched up to 94% correlation with manual results and reduced the time by 90%.
Analysis of feature importance in deep neural networks in psychiatric disorders using magnetic resonance imaging
Sánchez, I., Soriano-Mas, C., Verdejo-García, A., Cardoner, N., Fernández-Aranda, F., Menchón, J. M., Rodrigues, P., Prčkovska, V., & Rowe, M. (2019). Analysis of feature importance in deep neural networks in psychiatric disorders using magnetic resonance imaging. In ISMRM 2019: 27th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
In this work, we train a neural network to differentiate between healthy subjects and patients of six different mental illnesses and we analyzed the network weights of the model to identify the most important regions of the brain for classification.
Reproducibility of SIENAX volumetric outputs over intra-session, inter-session and inter-scanner acquisitions
García, G., Moreno-Dominguez, D., Rowe, M., Prčkovska, V., & Rodrigues, P. (2019). Reproducibility of SIENAX volumetric outputs over intra-session, inter-session and inter-scanner acquisitions. In ISMRM 2019: 27th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
We conducted a reliability analysis for SIENAX in a test-retest dataset and a multi-site dataset. The volumetric outputs of SIENAX show low coefficients of variance for the test-retest dataset but quite higher multi-site data, suggesting a possible need for data harmonization in multi-site studies.
2018
Automated signal intensity quantification software – initial real world clinical validation
Wang, J., Moser, F., Moreno-Dominguez, D., Ramos, M., Prčkovska, V., Rodrigues, P., Markus Blank, S., & Agarwal, J. A. (2018). Automated signal intensity quantification software – initial “real world” clinical validation. Presentation at Western Neuroradiological Society 49th Annual Meeting.
Extensive research is now underway in trying to understand the mechanism of increased signal intensity (SI) and gadolinium presence in the brain and whether it has a clinical implication for patients. in this work, automated signal intensity quantification software was developed and validated in patient data from clinical routine.
Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge
Bakas, S., Reyes, M., et int. (…, Puch, S., Sánchez, I., Prc̆kovska, V., …), & Menze, B. (2018). Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge.
This study assesses the state-of-the-art machine learning methods used for brain tumor image analysis in MRI scans during the last instances of the Brain Tumor Segmentation (BraTS) challenge.
Multiple Sclerosis Lesion Segmentation using Improved Convolutional Neural Networks.
Kazancli, E., Prchkovska, V., Rodrigues, P., Villoslada, P., & Igual, L. (2018). Multiple Sclerosis Lesion Segmentation using Improved Convolutional Neural Networks. In Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (pp. 260–269).
This publication evaluates novel approaches to segment Multiple Sclerosis lesions from MR scans based on Convolutional Neural Networks.
Cross-vendor and Cross-protocol harmonisation of diffusion MRI data: a comparative study
Tax, C. M. W., Grussu, F., et int. (…, Puch, S., Rowe, M., Rodrigues, P., Prčkovska, V., …) &Veraart, J. (2018). Cross-vendor and Cross-protocol harmonization of diffusion {MRI} data: a comparative study. In ISMRM 2018: 26th Annual Meeting of the International Society for Magnetic Resonance in Medicine (p. 471).
This paper presents the diffusion MRI harmonisation challenge held at MICCAI 2017, in which five different methods that estimate mappings between scanners for diffusion MRI data harmonisation were evaluated on a dedicated dataset of the same subjects acquired on three distinct scanners.
Classification of subjects with psychiatric disorders using Deep Learning and identification of relevant features in the data
Sánchez, I. (2018). Classification of subjects with disorders using Deep Learning and identification of relevant features in the data. Universitat Politècnica de Catalunya.
In this Thesis, a deep neural network was trained to classify between healthy control subjects and subjects suffering from six different mental illnesses. By pre-processing T1 images and rs-fMRI, morphological changes in terms of the volume of brain regions, and changes in functionality between these regions were used as input data. Using the trained weights of the model and a novel visualization tool implemented during the course of the Thesis, it was studied which regions of the brain can be used as potential biomarkers for improving the diagnosis of brain disorders.
Multimodal brain tumor segmentation in Magnetic Resonance Images with Deep Architectures
Puch, S. (2018). Multimodal brain tumor segmentation in Magnetic Resonance Images with Deep Architectures. Universitat Autònoma de Barcelona.
In this thesis, we evaluate multiple 3D convolutional neural networks on the task of brain tumor segmentation. We focus on context-aware and efficient architectures, and we train these architectures in two large datasets, the Brain Tumor Segmentation Challenge (BraTS) dataset and the QMENTA
Brain Tumors dataset.
2017
ISMRM 2017: TraCED Challenge Entry
Rowe, M., Rodrigues, P., & Prčkovska, V. (2017). ISMRM 2017: TraCED Challenge Entry – 3rd place. In ISMRM 2017: TraCED Challenge (pp. 2–4).
In this work, we present a methodology for automatically segmenting multiple white matter fiber structures using a combination of white matter ROIs and T1-weighted-derived gray matter parcellation. The automated method segments the fascicle structures defined in the challenge without any manual intervention.
MultipleMS: A distributed workflow for managing and processing neuroimaging data
Peeters, T.H.J.M., Lazovski, N., Puch, S., Alkin, A., Moreno-Dominguez, D., & Prčkovska, V. (2017). MultipleMS: A distributed workflow for managing and processing neuroimaging data.
In this whitepaper, we describe one such neuroscience project and we present our solution at QMENTA implemented together with the UCSF School of Medicine, Department of Neurology. In the end, we summarize the benefits of our approach and inform you on how you can also use our platform with minimum effort.
Data security on QMENTA cloud: security frameworks on the platform.
Sato, T., Alkin, A., Lazovski, N., & Rodrigues, P. (2017). Data security on QMENTA cloud: security frameworks on the platform.
QMENTA understands the security requirements of the cloud architecture and the platform is designed to deliver even better security than traditional on-premises data archiving systems. Our comprehensive security strategy includes technical implementation, organizational structure, and many other business operations. The client’s data is protected at all times – whether it is travelling over the internet or stored on the platform.
Neuroimaging workflow in the cloud: standardizing research
Lazovski, N., Ramos, M., Moreno-Dominguez, D., Sato, T., Peeters, T., Prčkovska, V., & Rodrigues, P. (2017). Neuroimaging workflow in the cloud : standardizing research. In OHBM 2017: 23rd Annual Meeting of the Organization for Human Brain Mapping.
Neuroscientists often face a critical problem hindering collaboration and reproducible research: lack of an efficient and standardized way to share data and to share analytic tools. In this work, we propose a web-based cloud system CloudN designed for neuroimaging workflow, enabling storage, quality control, version control, sharing, analysis, and visualization of various aspects of the neuroimaging data.
BrainVis: A cloud-connected 3D exploration and visualization tool for multi-modal neuroimaging data
Prčkovska, V., Peeters, T., Moreno-Dominguez, D., & Rodrigues, P. (2017). BrainVis: A cloud-connected 3D exploration and visualization tool for multi-modal neuroimaging data. In OHBM 2017: 23rd Annual Meeting of the Organization for Human Brain Mapping.
in ISMRM 2017: 25th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
There is a lack of standardization in the current neuroimaging viewers, with various file types and data structures. We developed BrainVis to alleviate these problems: it is a fully interactive 3D viewer, where multiple modalities can be shown simultaneously. Advanced tools are provided such as surface rendering and interactive exploration of tractography streamlines. Seamless integration with a cloud-system provides transparent fetching and processing of data.
Wired minds: The neural underpinnings of the entrepreneurial brain
Rodrigues, P. R., Moreno-Dominguez, D., Ramos, M., Villoslada, P., Gallardo-pujol, D., & Prčkovska, V. (2017). minds: The neural underpinning of the entrepreneurial brain.
In OHBM 2017: 23rd Annual Meeting of the Organization for Human Brain Mapping.
in ISMRM 2017: 25th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
Neural underpinnings of entrepreneurship are not well understood yet. Recent publications suggest that the behavior of psychopaths and entrepreneurs is not very different. This study aims to take an overall view of personality traits typically associated with entrepreneurship and link them to connectivity indices and cortical brain measurements.
(free)Surfing ANTs: a comparative study
Puch, S., Rodrigues, P., Moreno-Dominguez, D., Ramos, M., & Prčkovska, V. (2017). (free)Surfing ANTs: a comparative study. In ISMRM 2017: 25th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
In this work, we analyze the reproducibility and repeatability of two common brain segmentation tools: FreeSurfer and ANTs, and examine their differences for various brain structures.
Multiple sclerosis lesion segmentation using deep learning
Kazancli, E. (2017). Multiple Sclerosis Lesion Segmentation using Deep Learning. Universitat Politecnica de Catalunya.
In this thesis, we evaluate novel approaches to segment Multiple Sclerosis lesions from MR scans based on Convolutional Neural Networks.
2016
Wired minds: How personality traits can predict entrepreneurs’ brains
Ledezma-Haight, R., Ramos, M., Prčkovska, V., Rodrigues, P., & Gallardo-Pujol, D. (2016). Wired minds: How personality traits can predict entrepreneurs’ brains. Personality and Individual Differences, 101, 493.
Very little is known on how the entrepreneurial brain works and what are the differences. This work aims to take an overall view of the traits found in an entrepreneur (determined by psychometric assessments) and compare these to connectivity levels and volumes in certain areas of the brain.
Characterizing functional connectivity during rest in multiple sclerosis patients versus healthy volunteers using independent component analysis.
Palacio, L. (2016). Characterizing functional connectivity during rest in multiple sclerosis patients versus healthy volunteers using independent component analysis. Universitat Pompeu Fabra.
In order to study the effects of multiple sclerosis on the functional connectivity of the brain, we applied a numerical method known as independent component analysis (ICA) and implemented a web user interface to allow the user to manually classify all the independent components for a given subject. We did not find any significant functional connectivity differences between MS patients and healthy volunteers.
2015
Contextual Diffusion Image Post-processing Aids Clinical Applications
Prčkovska, V., Andorrà, M., Villoslada, P., Martinez-Heras, E., Duits, R., Fortin, D., … Descoteaux, M. (2015). Contextual Diffusion Image Post-processing Aids Clinical Applications. Mathematics and Visualization, 40(1), 353–377.
In this paper, we present a possibility in enabling HARDI tractography on the data acquired under limited diffusion tensor imaging conditions. We enhance local features from the tensor field taking ‘context’ information into account. Moreover, we demonstrate the potential of the contextual processing techniques in two important clinical applications: enhancing the streamlines in data acquired from patients with Multiple Sclerosis (MS) and pre-surgical planning for tumor resection
Reproducibility of the Structural Connectome Reconstruction across Diffusion Methods
Prčkovska, V., Rodrigues, P., Puigdellivol Sanchez, A., Ramos, M., Andorra, M., Martinez-Heras, E., … Villoslada, P. (2015). Reproducibility of the Structural Connectome Reconstruction across Diffusion Methods. Journal of Neuroimaging, 26(1), 46–57.
In this work, we evaluated the reproducibility of structural connectome techniques on test-retest and longitudinal data from 22 healthy volunteers. We compared connectivity matrices and tract reconstructions obtained with the most typical acquisition schemes used in clinical application.
The effect of the resampling in DTI tractograms
Ramos, M., Serret, C., Tudela, R., Rodrigues, P., Falcón, C., Soria, G., & Prchkovska, V. (2015). The effect of the resampling in DTI tractograms. ISMRM 2015: 6th Annual Meeting of the Italian Chapter of the International Society for Magnetic Resonance in Medicine.
In this work, we investigate the effect of 3 interpolation techniques (trilinear, nearest neighbor and Fischer’s Bresenham interpolation) when resampling to an isotropic resolution on the estimation of the derived DTI scalar maps and reconstructed fibers on “in-vivo” mice data. Results show significant changes in FA values along fibers.
What does it take to build a qMRI laboratory?
Rodrigues, P. (2015). What does it take to build a qMRI laboratory? ISMRM 2015: 6th Annual Meeting of the Italian Chapter of the International Society for Magnetic Resonance in Medicine.
An overview through our journey building QMENTA from an idea to an award-winning start-up, and the vision and goals behind it.
Web-cloud platform for storing, processing and analyzing multi-modal neuroimaging data
Lazovski, N., Ramos, M., & Rodrigues, P. (2015). Web-cloud platform for storing, processing and analyzing multi-modal neuroimaging data. ISMRM 2015: 6th Annual Meeting of the Italian Chapter of the International Society for Magnetic Resonance in Medicine.
In this work, we present one of our first designs for the QMENTA web-based cloud system used to store, process, analyze, and visualize aspects of the neuroimaging muli-modal data.
Improved analysis framework for MS connectomes
Andorra, M., Ramos, M., Martinez-Heras, E., Lampert, E., Rodrigues, P., Villoslada, P., & Prchkovska, V. (2015). Improved analysis framework for MS connectomes. ISMRM 2015: 6th Annual Meeting of the Italian Chapter of the International Society for Magnetic Resonance in Medicine.
In this work, we found that HARDI acquisition showed the most balanced trade-off between high reproducibility of the connectome, higher rate of path detection and of fanning fibers, and intermediate acquisition times (10-15 minutes), although at the cost of a higher appearance of aberrant fibers.
Reproducibility of the structural connectome and other open challenges
Rodrigues, P., Prats-Galino, A., Gallardo-Pujol, D., Villoslada, P., Falcon, C., & Prčkovska, V. (2014). Reproducibility of the structural connectome and other open challenges. In ISMRM 2014: 22th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
In this work we address the reproducibility of the structural connectome if acquired under different q-space sampling conditions, and in the same and different scanning session. We also address several caveats that should be taken with care when performing this type of analysis approaches with DWI data.
Evaluating Structural Connectomics: the effect of the cortical parcellation scheme
Rodrigues, P., Prats-Galino, A., Gallardo-Pujol, D., Villoslada, P., Prčkovska, V., & Falcon, C. (2013). Evaluating Structural Connectomics: the effect of the cortical parcellation scheme. Frontiers in Neuroinformatics, 7(May 2015), 29–32.
In this work, we evaluated the information difference contained in the structural connectomes constructed from the same subject, at different levels of the cortical parcellation, scanned with different dMRI acquisition techniques (DTI, HARDI and DSI).
2014
Reproducibility of the structural connectome and other open challenges
Rodrigues, P., Prats-Galino, A., Gallardo-Pujol, D., Villoslada, P., Falcon, C., & Prčkovska, V. (2014). Reproducibility of the structural connectome and other open challenges. In ISMRM 2014: 22th Annual Meeting of the International Society for Magnetic Resonance in Medicine.
In this work we address the reproducibility of the structural connectome if acquired under different q-space sampling conditions, and in the same and different scanning session. We also address several caveats that should be taken with care when performing this type of analysis approaches with DWI data.
2013
Evaluating Structural Connectomics: the effect of the cortical parcellation scheme
Rodrigues, P., Prats-Galino, A., Gallardo-Pujol, D., Villoslada, P., Prčkovska, V., & Falcon, C. (2013). Evaluating Structural Connectomics: the effect of the cortical parcellation scheme. Frontiers in Neuroinformatics, 7(May 2015), 29–32.
In this work, we evaluated the information difference contained in the structural connectomes constructed from the same subject, at different levels of the cortical parcellation, scanned with different dMRI acquisition techniques (DTI, HARDI and DSI).
Evaluating Structural Connectomics in Relation to Different Q-space Sampling Techniques
Rodrigues, P., Prats-Galino, A., Gallardo-Pujol, D., Villoslada, P., Falcon, C., & Prčkovska, V. (2013). Evaluating structural connectomics in relation to different Q-space sampling techniques. Lecture Notes in Computer Science (Including Lecture Notes in Bioinformatics), 8149 LNCS(PART 1), 671–678.
In this work, we evaluate the structural connectome by analyzing and comparing graph-based measures on real data acquired using the three most important Diffusion Weighted Imaging techniques: DTI, HARDI and DSI.
Diffusion Tensor Imaging study for Multiple Sclerosis in the Optic Radiation
Tehrani, M. A., Martinez-Heras, E., Rodrigues, P., Gabilondo, I., Falcon, C., Villoslada, P., & Prckovska, V. (2013). Diffusion Tensor Imaging study for Multiple Sclerosis in the Optic Radiation. In ISMRM Workshop on Multiple Sclerosis as a Whole-Brain Disease.
Diffusion tensor imaging (DTI) enables the reconstruction of fiber bundles and white matter tracts and previous findings have demonstrated that it is sensitive to the evolution of tissue damage within Multiple Sclerosis lesions. We investigate the sensitivity power of scalar indices derived from DTI, for detecting lesions present in the white matter of Multiple Sclerosis patients.