1
|
Musall BC, Gabr RE, Yang Y, Kamali A, Lincoln JA, Jacobs MA, Ly V, Luo X, Wolinsky JS, Narayana PA, Hasan KM. Detection of diffusely abnormal white matter in multiple sclerosis on multiparametric brain MRI using semi-supervised deep learning. Sci Rep 2024; 14:17157. [PMID: 39060426 PMCID: PMC11282266 DOI: 10.1038/s41598-024-67722-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/22/2024] [Accepted: 07/15/2024] [Indexed: 07/28/2024] Open
Abstract
In addition to focal lesions, diffusely abnormal white matter (DAWM) is seen on brain MRI of multiple sclerosis (MS) patients and may represent early or distinct disease processes. The role of MRI-observed DAWM is understudied due to a lack of automated assessment methods. Supervised deep learning (DL) methods are highly capable in this domain, but require large sets of labeled data. To overcome this challenge, a DL-based network (DAWM-Net) was trained using semi-supervised learning on a limited set of labeled data for segmentation of DAWM, focal lesions, and normal-appearing brain tissues on multiparametric MRI. DAWM-Net segmentation performance was compared to a previous intensity thresholding-based method on an independent test set from expert consensus (N = 25). Segmentation overlap by Dice Similarity Coefficient (DSC) and Spearman correlation of DAWM volumes were assessed. DAWM-Net showed DSC > 0.93 for normal-appearing brain tissues and DSC > 0.81 for focal lesions. For DAWM-Net, the DAWM DSC was 0.49 ± 0.12 with a moderate volume correlation (ρ = 0.52, p < 0.01). The previous method showed lower DAWM DSC of 0.26 ± 0.08 and lacked a significant volume correlation (ρ = 0.23, p = 0.27). These results demonstrate the feasibility of DL-based DAWM auto-segmentation with semi-supervised learning. This tool may facilitate future investigation of the role of DAWM in MS.
Collapse
Affiliation(s)
- Benjamin C Musall
- Department of Diagnostic and Interventional Imaging, University of Texas McGovern Medical School, 6431 Fannin St., MSE 168, Houston, TX, 77030, USA
| | - Refaat E Gabr
- Department of Diagnostic and Interventional Imaging, University of Texas McGovern Medical School, 6431 Fannin St., MSE 168, Houston, TX, 77030, USA
| | - Yanyu Yang
- Department of Biostatistics and Data Science, University of Texas School of Public Health, Houston, TX, USA
| | - Arash Kamali
- Department of Diagnostic and Interventional Imaging, University of Texas McGovern Medical School, 6431 Fannin St., MSE 168, Houston, TX, 77030, USA
| | - John A Lincoln
- Department of Neurology, University of Texas McGovern Medical School, Houston, TX, USA
| | - Michael A Jacobs
- Department of Diagnostic and Interventional Imaging, University of Texas McGovern Medical School, 6431 Fannin St., MSE 168, Houston, TX, 77030, USA
- The Russell H. Morgan Department of Radiology and Radiological Science and Sidney Kimmel Comprehensive Cancer Center, The Johns Hopkins School of Medicine, Baltimore, MD, USA
- Department of Computer Science, Rice University, Houston, TX, USA
| | - Vi Ly
- Department of Biostatistics and Data Science, University of Texas School of Public Health, Houston, TX, USA
| | - Xi Luo
- Department of Biostatistics and Data Science, University of Texas School of Public Health, Houston, TX, USA
| | - Jerry S Wolinsky
- Department of Neurology, University of Texas McGovern Medical School, Houston, TX, USA
| | - Ponnada A Narayana
- Department of Diagnostic and Interventional Imaging, University of Texas McGovern Medical School, 6431 Fannin St., MSE 168, Houston, TX, 77030, USA
| | - Khader M Hasan
- Department of Diagnostic and Interventional Imaging, University of Texas McGovern Medical School, 6431 Fannin St., MSE 168, Houston, TX, 77030, USA.
| |
Collapse
|
2
|
Rovira À, Llufriu S. Coordination between neuroradiology and neurology departments in the care of patients with multiple sclerosis: Recommendations for optimization. RADIOLOGIA 2022; 64:379-382. [DOI: 10.1016/j.rxeng.2021.09.011] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/06/2021] [Accepted: 09/07/2021] [Indexed: 11/25/2022]
|
3
|
Rovira À, Llufriu S. Coordinación de los servicios de neurorradiología y neurología en la atención a pacientes con esclerosis múltiple: recomendaciones para su optimización. RADIOLOGIA 2022. [DOI: 10.1016/j.rx.2021.09.005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/26/2022]
|
4
|
Winn A, Martin A, Castellon I, Sanchez A, Lavi ES, Munera F, Nunez D. Spine MRI: A Review of Commonly Encountered Emergent Conditions. Top Magn Reson Imaging 2021; 29:291-320. [PMID: 33264271 DOI: 10.1097/rmr.0000000000000261] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/25/2022]
Abstract
Over the last 2 decades, the proliferation of magnetic resonance imaging (MRI) availability and continuous improvements in acquisition speeds have led to significantly increased MRI utilization across the health care system, and MRI studies are increasingly ordered in the emergent setting. Depending on the clinical presentation, MRI can yield vital diagnostic information not detectable with other imaging modalities. The aim of this text is to report on the up-to-date indications for MRI of the spine in the ED, and review the various MRI appearances of commonly encountered acute spine pathology, including traumatic injuries, acute non traumatic myelopathy, infection, neoplasia, degenerative disc disease, and postoperative complications. Imaging review will focus on the aspects of the disease process that are not readily resolved with other modalities.
Collapse
Affiliation(s)
- Aaron Winn
- University of Miami, Jackson Memorial Hospital, Miami, FL
| | - Adam Martin
- University of Miami, Jackson Memorial Hospital, Miami, FL
| | - Ivan Castellon
- University of Miami, Jackson Memorial Hospital, Miami, FL
| | - Allen Sanchez
- University of Miami, Jackson Memorial Hospital, Miami, FL
| | | | - Felipe Munera
- University of Miami, Jackson Memorial Hospital, Miami, FL
| | - Diego Nunez
- Brigham and Women's Hospital, Harvard Medical School, Boston, MA
| |
Collapse
|
5
|
Dahlqvist JR, Widholm P, Leinhard OD, Vissing J. MRI in Neuromuscular Diseases: An Emerging Diagnostic Tool and Biomarker for Prognosis and Efficacy. Ann Neurol 2020; 88:669-681. [PMID: 32495452 DOI: 10.1002/ana.25804] [Citation(s) in RCA: 46] [Impact Index Per Article: 11.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/12/2020] [Revised: 05/05/2020] [Accepted: 05/25/2020] [Indexed: 12/12/2022]
Abstract
There is an unmet need to identify biomarkers sensitive to change in rare, slowly progressive neuromuscular diseases. Quantitative magnetic resonance imaging (MRI) of muscle may offer this opportunity, as it is noninvasive and can be carried out almost independent of patient cooperation and disease severity. Muscle fat content correlates with muscle function in neuromuscular diseases, and changes in fat content precede changes in function, which suggests that muscle MRI is a strong biomarker candidate to predict prognosis and treatment efficacy. In this paper, we review the evidence suggesting that muscle MRI may be an important biomarker for diagnosis and to monitor change in disease severity. ANN NEUROL 2020;88:669-681.
Collapse
Affiliation(s)
- Julia R Dahlqvist
- Copenhagen Neuromuscular Center, Department of Neurology, Rigshospitalet, Copenhagen University, Copenhagen, Denmark
| | - Per Widholm
- Department of Medical and Health Sciences, Linköping University, Linköping, Sweden
- AMRA Medical AB, Linköping, Sweden
| | - Olof Dahlqvist Leinhard
- Department of Medical and Health Sciences, Linköping University, Linköping, Sweden
- AMRA Medical AB, Linköping, Sweden
| | - John Vissing
- Copenhagen Neuromuscular Center, Department of Neurology, Rigshospitalet, Copenhagen University, Copenhagen, Denmark
| |
Collapse
|
6
|
Zaccagna F, Matys T, Massoud TF. Optic Chiasm Morphometric Changes in Multiple Sclerosis: Feasibility of a Simplified Brain Magnetic Resonance Imaging Measure of White Matter Atrophy. Clin Anat 2019; 32:1072-1081. [PMID: 31381196 DOI: 10.1002/ca.23446] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/10/2019] [Accepted: 08/01/2019] [Indexed: 11/11/2022]
Abstract
Sophisticated volume measurements of brain structures on magnetic resonance imaging (MRI) may improve specificity in determining long-term progression of multiple sclerosis (MS), but these techniques are laborious. The optic chiasm (OC) is a white matter (WM) structure clearly visible on a routine MRI and is related to the optic nerves (ONs), which are known to atrophy in MS. We hypothesized that OC morphometric measurements would show OC atrophy in MS compared to normal patients. If so, this could help establish a novel simplified brain MRI measure of WM atrophy in MS patients. We retrospectively evaluated standard brain MRIs of 97 patients with known MS and 98 normal individuals. We electronically measured eight OC morphometrics on axial T2WIs and midsagittal T1WIs: OC width and anteroposterior (AP) diameter, diameters of each ON and optic tract (OT), and angles between the ONs or OTs. Mean OC width, AP diameter, and height in MS patients were 11.83 ± 1.25 mm (95% CI 11.58-12.09), 2.99 ± 0.65 mm (95% CI 2.85-3.12), and 2.09 ± 0.37 mm (95% CI 2-2.19), respectively. In normal individuals, they were 12.1 ± 1.4 mm (95% CI 11.78-12.34), 3.43 ± 0.63 mm (95% CI 3.3-3.58), and 2.15 ± 0.37 mm (95% CI 2.07-2.23), respectively. There were statistically significant differences between MS patients and controls for AP diameter (P = 0.000), but not for width (P = 0.204) or height (P = 0.183). The ONs were significantly smaller in MS (P < 0.0017), but not the OTs. Thus, the OC is significantly atrophied in an unstratified cohort of MS patients. Future studies may establish an MRI OC morphometric index to evaluate demyelinating disease in the brain. Clin. Anat. 32:1072-1081, 2019. © 2019 Wiley Periodicals, Inc.
Collapse
Affiliation(s)
- Fulvio Zaccagna
- Section of Neuroradiology, Department of Radiology, University of Cambridge School of Clinical Medicine, Cambridge, UK.,Division of Neuroimaging and Neurointervention, Department of Radiology, Stanford University School of Medicine, Stanford, California
| | - Tomasz Matys
- Section of Neuroradiology, Department of Radiology, University of Cambridge School of Clinical Medicine, Cambridge, UK.,Division of Neuroimaging and Neurointervention, Department of Radiology, Stanford University School of Medicine, Stanford, California
| | - Tarik F Massoud
- Section of Neuroradiology, Department of Radiology, University of Cambridge School of Clinical Medicine, Cambridge, UK.,Division of Neuroimaging and Neurointervention, Department of Radiology, Stanford University School of Medicine, Stanford, California
| |
Collapse
|
7
|
Quantitative magnetic resonance assessment of brain atrophy related to selected aspects of disability in patients with multiple sclerosis: preliminary results. Pol J Radiol 2019; 84:e171-e178. [PMID: 31481987 PMCID: PMC6717938 DOI: 10.5114/pjr.2019.84274] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/21/2018] [Accepted: 03/25/2018] [Indexed: 12/01/2022] Open
Abstract
Purpose The aim of this volumetric study was to evaluate the relationship between brain atrophy quantification in multiple sclerosis (MS) patients and the progression of disability measured by neurological standardised tests. Material and methods Seventeen patients (mean age 40.89 years) with clinically definite MS and 24 control subjects (mean age 38.45 years) were enrolled in the study. Brain examinations were performed on a 1.5T MR scanner. Automatic brain segmentation was done using FreeSurfer. Neurological disability was assessed in all patients in baseline and after a median follow-up of two years, using EDSS score evaluation. Results In MS patients we found significantly (p < 0.05) higher atrophy rates in many brain areas compared with the control group. The white matter did not show any significant rate of volume loss in MS patients compared to healthy controls. Significant changes were found only in grey matter volume in MS subjects. At the follow-up evaluation after two years MS patients with deterioration in disability revealed significantly decreased cerebral volume in 14 grey matter areas at baseline magnetic resonance imaging (MRI) compared to MS subjects without disability progression. Conclusions Grey matter atrophy is associated with the degree of disability in MS patients. Our results suggest that morphometric measurements of brain volume could be a promising non-invasive biomarker in assessing the volumetric changes in MS patients as related to disability progression in the course of the disease.
Collapse
|
8
|
A Review on a Deep Learning Perspective in Brain Cancer Classification. Cancers (Basel) 2019; 11:cancers11010111. [PMID: 30669406 PMCID: PMC6356431 DOI: 10.3390/cancers11010111] [Citation(s) in RCA: 145] [Impact Index Per Article: 29.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/29/2018] [Revised: 01/07/2019] [Accepted: 01/10/2019] [Indexed: 12/12/2022] Open
Abstract
A World Health Organization (WHO) Feb 2018 report has recently shown that mortality rate due to brain or central nervous system (CNS) cancer is the highest in the Asian continent. It is of critical importance that cancer be detected earlier so that many of these lives can be saved. Cancer grading is an important aspect for targeted therapy. As cancer diagnosis is highly invasive, time consuming and expensive, there is an immediate requirement to develop a non-invasive, cost-effective and efficient tools for brain cancer characterization and grade estimation. Brain scans using magnetic resonance imaging (MRI), computed tomography (CT), as well as other imaging modalities, are fast and safer methods for tumor detection. In this paper, we tried to summarize the pathophysiology of brain cancer, imaging modalities of brain cancer and automatic computer assisted methods for brain cancer characterization in a machine and deep learning paradigm. Another objective of this paper is to find the current issues in existing engineering methods and also project a future paradigm. Further, we have highlighted the relationship between brain cancer and other brain disorders like stroke, Alzheimer’s, Parkinson’s, and Wilson’s disease, leukoriaosis, and other neurological disorders in the context of machine learning and the deep learning paradigm.
Collapse
|