PresentationPDF Available

Semi-automatic cloud-based workflow for evaluating the central vein sign for MS diagnosis in a multicenter clinical setting

Authors:

Abstract and Figures

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.
Content may be subject to copyright.
1
Semi-automatic cloud-based workflow for
evaluating the central vein sign for MS
diagnosis in a multicenter clinical setting
David Moreno-Domínguez 1, Marc Ramos 1,
Daniel S. Reich 2, Daniel Ontaneda 3,
Paulo Rodrigues 1, Pascal Sati 2
1 QMENTA Inc, 2 National Institutes of Health , 3 Cleveland Clinic
2019 - Marc Ramos - QMENTA - marc@qmenta.com 2
Speaker Name: David Moreno-Dominguez
I have the following financial interest or relationship to disclose with
regard to the subject matter of this presentation:
Company Name: QMENTA
Type of Relationship: Employee and stock options holder
Declaration of
Financial Interests or Relationships
2019 - David Moreno-Dominguez - QMENTA® - david@qmenta.com 3
Introduction
Previous work
The central vein sign (CVS) has
recently been proposed as a novel
MRI marker for improving the
accuracy and reducing the time to
diagnose patients with multiple
sclerosis (MS)1.
Motivation
CVS has been found in a majority of MRI white matter
lesions in patients with MS and in significantly lower
numbers in other conditions associated with white
matter lesions. However, to determine if CVS will
improve diagnostic accuracy for MS in clinical practice,
a prospective multi-center study of CVS in patients
undergoing new evaluation for suspect MS is needed.
1. Sati et al. The central vein sign and its clinical evaluation for the diagnosis of multiple sclerosis. Nature Reviews Neurology. 2016 Dec;12(12):714.
2. Sati et al. FLAIR*: a combined MR contrast technique for visualizing white matter lesions and parenchymal veins. Radiology. 2012 Dec;265(3):926-32.
3. Sati et al. Rapid MR susceptibility imaging of the brain using segmented 3D echo-planar imaging (3D EPI) and its clinical applications. Magnetom FLASH. 2017;68:26-32.
Our approach
In this work, we developed a semi-automatic
cloud-based workflow for evaluating the clinical value
of the CVS for MS diagnosis using FLAIR* in a
multicenter setting.
Recent advancements in the form of novel MRI
techniques have introduced the FLAIR* contrast,
combining submillimeter resolution and
improved contrast for lesions and veins,
allowing for the central vein sign to be easily
identified 2,3.
2019 - David Moreno-Dominguez - QMENTA® - david@qmenta.com 4
Data
The study aims at recruiting 10 participants at each of 10 North American Imaging in MS Cooperative
(NAIMS) sites. Referred to an MS center for a possible diagnosis of MS based on a clinical or radiological
suspicion.
3T MRI protocol includes:
3D T2-FLAIR (1 mm isotropic, TR = 4.8 s, TE = 352 ms. TI = 1800 ms)
T2*-weighted 3D-EPI (0.65 mm isotropic, TR = 64 ms, TE = 35 ms. EPI factor = 15)
acquired pre- and post-injection of gadolinium agent.
DICOM images are uploaded and automatically anonymized on a cloud-based platform accessible
through a web browser.
Image types are automatically
recognized and tagged, allowing for
automatic processing without file
selection.
2019 - David Moreno-Dominguez - QMENTA® - david@qmenta.com 5
Cloud setup
Project Owner
Project Lead
(NIH/NINDS)
Cloud Image Management and
Analysis system
Scanning site
(Cleveland Clinic)
Rater
Technician
Scanning site
Rater
Technician
Scanning site
Rater
Technician
2019 - David Moreno-Dominguez - QMENTA® - david@qmenta.com 6
Workflow
A custom- built neuroimaging analysis workflow was implemented in the cloud platform including the
following processing steps:
calculation of FLAIR* images
Automated lesion segmentation and central image / segmentation manual quality assessment
Interactive manual annotation of the CVS: The presence of central veins will be determined by
raters at each site using guidelines published by the NAIMS Cooperative.
Centralized percentage-based rating of the CVS in each participant.
2019 - David Moreno-Dominguez - QMENTA® - david@qmenta.com 7
Automatic processing
2019 - David Moreno-Dominguez - QMENTA® - david@qmenta.com 8
Assisted Manual Steps: Lesion QC
Automatic Lesion Segmentation
Browser-based
Semi-automatic lesion QC
Full interactive list of detected lesions
Reject false positives with one click
Add false negatives with one click
Optional manual “paintbrush” editing
2019 - David Moreno-Dominguez - QMENTA® - david@qmenta.com 9
Assisted Manual Steps: CV Annotation
Browser-based CVS annotation
Zoom, Pan and Transparency control
Annotate CV+ lesions with one click
Center image on annotation by clicking on
entry
Automatic count of CV+ at the end of the
session
2019 - David Moreno-Dominguez - QMENTA® - david@qmenta.com 10
Discussion
This ongoing pilot prospective study is aimed at demonstrating the technical feasibility of
robustly imaging the CVS in a multicenter setting, and providing additional evidence of the
potential added value of the CVS for the diagnosis of MS.
The use of a cloud-based platform in this study allows to easily collect and secure data
across sites, while simplifying the workflow of neuroimaging analysis.
This novel workflow is a powerful tool that has the potential to significantly accelerate the
clinical research imaging studies in MS.
11
Thank you for your
attention
David Moreno-Dominguez
Contact
david@qmenta.com
Booth 1010
ResearchGate has not been able to resolve any citations for this publication.
Poster
Full-text available
Valuable time and funding are spent on gathering fragmented and dispersedly stored data, and researchers are distracted from their main focus: understanding the brain and how disease affects it, and developing new neuroimaging tools to help in this endeavour. Data organization and solid analytics tools are essential to generate accurate and reproducible results. However, 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. We propose a web-based cloud system CloudNdesigned for neuroimaging workflow, enabling storage, quality control, version control, sharing, analysis, and visualization of various aspects of the neuroimaging data.
Article
Full-text available
Over the past few years, MRI has become an indispensable tool for diagnosing multiple sclerosis (MS). However, the current MRI criteria for MS diagnosis have imperfect sensitivity and specificity. The central vein sign (CVS) has recently been proposed as a novel MRI biomarker to improve the accuracy and speed of MS diagnosis. Evidence indicates that the presence of the CVS in individual lesions can accurately differentiate MS from other diseases that mimic this condition. However, the predictive value of the CVS for the development of clinical MS in patients with suspected demyelinating disease is still unknown. Moreover, the lack of standardization for the definition and imaging of the CVS currently limits its clinical implementation and validation. On the basis of a thorough review of the existing literature on the CVS and the consensus opinion of the members of the North American Imaging in Multiple Sclerosis (NAIMS) Cooperative, this article provides statements and recommendations aimed at helping radiologists and neurologists to better understand, refine, standardize and evaluate the CVS in the diagnosis of MS.
Article
Tissue classification plays a crucial role in the investigation of normal neural development, brain-behavior relationships, and the disease mechanisms of many psychiatric and neurological illnesses. Ensuring the accuracy of tissue classification is important for quality research and, in particular, the translation of imaging biomarkers to clinical practice. Assessment with the human eye is vital to correct various errors inherent to all currently available segmentation algorithms. Manual quality assurance becomes methodologically difficult at a large scale - a problem of increasing importance as the number of data sets is on the rise. To make this process more efficient, we have developed Mindcontrol, an open-source web application for the collaborative quality control of neuroimaging processing outputs. The Mindcontrol platform consists of a dashboard to organize data, descriptive visualizations to explore the data, an imaging viewer, and an in-browser annotation and editing toolbox for data curation and quality control. Mindcontrol is flexible and can be configured for the outputs of any software package in any data organization structure. Example configurations for three large, open-source datasets are presented: the 1000 Functional Connectomes Project (FCP), the Consortium for Reliability and Reproducibility (CoRR), and the Autism Brain Imaging Data Exchange (ABIDE) Collection. These demo applications link descriptive quality control metrics, regional brain volumes, and thickness scalars to a 3D imaging viewer and editing module, resulting in an easy-to-implement quality control protocol that can be scaled for any size and complexity of study.
Article
Purpose: To evaluate a magnetic resonance (MR) imaging contrast technique, called FLAIR*, that combines the advantages of T2-weighted fluid-attenuated inversion recovery (FLAIR) contrast and T2*-weighted contrast on a single image for assessment of white matter (WM) diseases such as multiple sclerosis (MS). Materials and methods: This prospective pilot study was HIPAA compliant and institutional review board approved. Ten patients with clinically definite MS (eight men, two women; mean age, 41 years) provided informed consent and underwent 3.0-T MR imaging. Images from a T2-weighted FLAIR sequence were combined with images from a T2*-weighted segmented echo-planar imaging sequence performed during contrast material injection, yielding high-isotropic-resolution (0.55 × 0.55 × 0.55 mm(3)) FLAIR* images. Qualitative assessment was performed for image quality, lesion conspicuity, and vein conspicuity. Contrast-to-noise ratio (CNR) was calculated to compare normal-appearing WM (NAWM) with cerebrospinal fluid, lesions, and veins. To evaluate the differences in CNR among imaging modalities, a bootstrap procedure clustered on subjects was used, together with paired t tests. Results: High-quality FLAIR* images of the brain were produced at 3.0 T, yielding conspicuous lesions and veins. Lesion-to-NAWM and NAWM-to-vein CNR values were significantly higher for FLAIR* images than for T2-weighted FLAIR images (P < .0001). Findings on FLAIR* images included intralesional veins for lesions located throughout the brain and a hypointense rim around some WM lesions. Conclusion: High-isotropic-resolution FLAIR* images obtained at 3.0 T yield high contrast for WM lesions and parenchymal veins, making it well suited to investigate the relationship between WM abnormalities and veins in a clinical setting.
Article
One of the most challenging problems in modern neuroimaging is detailed characterization of neurodegeneration. Quantifying spatial and longitudinal atrophy patterns is an important component of this process. These spatiotemporal signals will aid in discriminating between related diseases, such as frontotemporal dementia (FTD) and Alzheimer's disease (AD), which manifest themselves in the same at-risk population. Here, we develop a novel symmetric image normalization method (SyN) for maximizing the cross-correlation within the space of diffeomorphic maps and provide the Euler-Lagrange equations necessary for this optimization. We then turn to a careful evaluation of our method. Our evaluation uses gold standard, human cortical segmentation to contrast SyN's performance with a related elastic method and with the standard ITK implementation of Thirion's Demons algorithm. The new method compares favorably with both approaches, in particular when the distance between the template brain and the target brain is large. We then report the correlation of volumes gained by algorithmic cortical labelings of FTD and control subjects with those gained by the manual rater. This comparison shows that, of the three methods tested, SyN's volume measurements are the most strongly correlated with volume measurements gained by expert labeling. This study indicates that SyN, with cross-correlation, is a reliable method for normalizing and making anatomical measurements in volumetric MRI of patients and at-risk elderly individuals.
Rapid MR susceptibility imaging of the brain using segmented 3D echo-planar imaging (3D EPI) and its clinical applications. Magnetom FLASH
  • P Sati
  • S Patil
  • S Inati
  • W T Wang
  • J A Derbyshire
  • G Krueger
  • D S Reich
  • J A Butman
Sati P, Patil S, Inati S, Wang WT, Derbyshire JA, Krueger G, Reich DS, Butman JA. Rapid MR susceptibility imaging of the brain using segmented 3D echo-planar imaging (3D EPI) and its clinical applications. Magnetom FLASH. 2017;68:26-32.
Bayesian inference for structured additive regression models for large-scale problems with applications to medical imaging (Doctoral dissertation, lmu)
  • P Schmidt
Schmidt P. Bayesian inference for structured additive regression models for large-scale problems with applications to medical imaging (Doctoral dissertation, lmu). URN: urn:nbn:de:bvb:19-203731