Combined brain and anterior visual pathways' MRIs assist in early identification of neuromyelitis optica spectrum disorder at onset of optic neuritis

Classification of optic neuritis in neuromyelitis optica spectrum disorders (NMOSD) on MRI using CNN with transfer learning and manipulation of pre-processing on augmentation

Neuromyelitis optica spectrum disorder (NMOSD), also known as Devic disease, is an autoimmune central nervous system disorder in humans that commonly causes inflammatory demyelination in the optic nerves and spinal cord. Inflammation in the optic nerves is termed optic neuritis (ON). ON is a common clinical presentation; however, it is not necessarily present in all NMOSD patients. ON in NMOSD can be relapsing and result in severe vision loss. To the best of our knowledge, no study utilises deep learning to classify ON changes on MRI among patients with NMOSD. Therefore, this study aims to deploy eight state-of-the-art CNN models (Inception-v3, Inception-ResNet-v2, ResNet-101, Xception, ShuffleNet, DenseNet-201, MobileNet-v2, and EfficientNet-B0) with transfer learning to classify NMOSD patients with and without chronic ON using optic nerve magnetic resonance imaging. This study also investigated the effects of data augmentation before and after dataset splitting on cropped and whole images. Both quantitative and qualitative assessments (with Grad-Cam) were used to evaluate the performances of the CNN models. The Inception-v3 was identified as the best CNN model for classifying ON among NMOSD patients, with accuracy of 99.5%, sensitivity of 98.9%, specificity of 93.0%, precision of 100%, NPV of 99.0%, and F1-score of 99.4%. This study also demonstrated that the application of augmentation after dataset splitting could avoid information leaking into the testing datasets, hence producing more realistic and reliable results.

Patterns of Adult Neuromyelitis Optica Spectrum Disorder Patients Compared to Multiple Sclerosis: A Systematic Review and Meta-Analysis

Neuromyelitis optica spectrum disorders (NMOSDs) are central nervous system inflammatory conditions, now recognized to involve the brain, often identified by aquaporin-4 (AQP4) antibodies. We aimed to summarize the characteristics of adult NMOSD patients compared to multiple sclerosis (MS). A computerized search was conducted on MEDLINE via PubMed, Web of Science, and ProQuest using the relevant keywords. Three independent reviewers performed two-stage screening and data extraction. The Review Manager 5.4 program (Cochrane Collaboration, Windows, London, UK) was used for the analysis. The Joanna Briggs Institute (JIB) tool was used for the quality of included studies. Twenty-three articles were included. NMOSD patients were associated with older age at presentation and higher Expanded Disability Status Scale (MD = 3.88, 95% CI: 1.80 to 5.97, P = 0.0003) and (MD = 1.15, 95% CI: 0.58 to 1.72, P < 0.0001), respectively. The risk of NMOSD in females was significantly higher than MS (OR = 2.21, 95% CI: 1.41 to 3.46, P = 0.0005). Patients with NMOSD were associated with a lower risk of extrapyramidal symptoms (OR = 0.26, 95% CI: 0.11 to 0.60, P < 0.01), brainstem involvement symptoms (OR = 0.32, 95% CI: 0.16 to 0.64, P < 0.01), and developing brain lesions compared to MS (OR = 0.08, 95% CI: 0.03 to 0.18, P < 0.00001). The current evidence suggests that both NMOSD and MS have different demographic, clinical, and lesion characteristics. There is a need for additional validation of the identified differences compared with MS due to the lack of long-term systematic imaging investigations in NMOSD.

Differences in the involved sites among different types of demyelinating optic neuritis in traditional MRI examination: A systematic review and meta-analysis

Background

Magnetic resonance imaging (MRI) plays a significant role in assessing optic neuropathy and providing more detailed information about the lesion of the visual pathway to help differentiate optic neuritis from other visual disorders. This study aims to systematically review the literature and verify if there is a real difference in lesion location among different demyelinating optic neuritis (DON) subtypes.

Methods

A systematic search was conducted including 8 electronic databases and related resources from the establishment of the database to August 25th, 2020. We classified DON into 5 subtypes and divided the visual pathways into five segments mainly comparing the differences in the involved visual pathway sites of different subtypes.

Results

Fifty-five studies were included in the analysis, and the abnormal rate was as high as 92% during the acute phase (within 4 weeks of symptom onset). With respect to lesion location, the orbital segment of the optic nerve was the most frequently involved (87%), whereas optic tract involvement was very rare. Involvement of the orbital segment was more common in myelin oligodendrocyte glycoprotein antibody-related optic neuritis (MOG-ON) (78%) and chronic relapsing inflammatory optic neuropathy (CRION) (81%), while the lesion was found to be located more posteriorly in neuromyelitis optica spectrum disorder-related optic neuritis (NMOSD-ON). With respect to lesion length, approximately 77% of MOG-ON patients had lesions involving more than half of the optic nerve length.

Conclusions

MRI examination is recommended for DON patients in the acute phase. In MOG-ON, anterior involvement is more common and the involved length is mostly more than 1/2 of the optic nerve length, whereas posterior involvement, intracranial segment, optic chiasm, or optic tract, is more common in NMOSD-ON.

Prospero registration number

CRD42020222430 (25-11-2020).

MRI Patterns Distinguish AQP4 Antibody Positive Neuromyelitis Optica Spectrum Disorder From Multiple Sclerosis

Neuromyelitis optica spectrum disorder (NMOSD) and multiple sclerosis (MS) are inflammatory diseases of the CNS. Overlap in the clinical and MRI features of NMOSD and MS means that distinguishing these conditions can be difficult. With the aim of evaluating the diagnostic utility of MRI features in distinguishing NMOSD from MS, we have conducted a cross-sectional analysis of imaging data and developed predictive models to distinguish the two conditions. NMOSD and MS MRI lesions were identified and defined through a literature search. Aquaporin-4 (AQP4) antibody positive NMOSD cases and age- and sex-matched MS cases were collected. MRI of orbits, brain and spine were reported by at least two blinded reviewers. MRI brain or spine was available for 166/168 (99%) of cases. Longitudinally extensive (OR = 203), "bright spotty" (OR = 93.8), whole (axial; OR = 57.8) or gadolinium (Gd) enhancing (OR = 28.6) spinal cord lesions, bilateral (OR = 31.3) or Gd-enhancing (OR = 15.4) optic nerve lesions, and nucleus tractus solitarius (OR = 19.2), periaqueductal (OR = 16.8) or hypothalamic (OR = 7.2) brain lesions were associated with NMOSD. Ovoid (OR = 0.029), Dawson's fingers (OR = 0.031), pyramidal corpus callosum (OR = 0.058), periventricular (OR = 0.136), temporal lobe (OR = 0.137) and T1 black holes (OR = 0.154) brain lesions were associated with MS. A score-based algorithm and a decision tree determined by machine learning accurately predicted more than 85% of both diagnoses using first available imaging alone. We have confirmed NMOSD and MS specific MRI features and combined these in predictive models that can accurately identify more than 85% of cases as either AQP4 seropositive NMOSD or MS.

Magnetic resonance imaging in neuromyelitis optica spectrum disorder

Neuromyelitis optica spectrum disorder (NMOSD) is an inflammatory disease of the central nervous system (CNS) associated with antibodies to aquaporin-4 (AQP4), which has distinct clinical, radiological and pathological features, but also has some overlap with multiple sclerosis and myelin oligodendrocyte glycoprotein (MOG) antibody associated disease. Early recognition of NMOSD is important because of differing responses to both acute and preventive therapy. Magnetic resonance (MR) imaging has proved essential in this process. Key MR imaging clues to the diagnosis of NMOSD are longitudinally extensive lesions of the optic nerve (more than half the length) and spinal cord (three or more vertebral segments), bilateral optic nerve lesions and lesions of the optic chiasm, area postrema, floor of the IV ventricle, periaqueductal grey matter, hypothalamus and walls of the III ventricle. Other NMOSD-specific lesions are denoted by their unique morphology: heterogeneous lesions of the corpus callosum, 'cloud-like' gadolinium (Gd)-enhancing white matter lesions and 'bright spotty' lesions of the spinal cord. Other lesions described in NMOSD, including linear periventricular peri-ependymal lesions and patch subcortical white matter lesions, may be less specific. The use of advanced MR imaging techniques is yielding further useful information regarding focal degeneration of the thalamus and optic radiation in NMOSD and suggests that paramagnetic rim patterns and changes in normal appearing white matter are specific to MS. MR imaging is crucial in the early recognition of NMOSD and in directing testing for AQP4 antibodies and guiding immediate acute treatment decisions. Increasingly, MR imaging is playing a role in diagnosing seronegative cases of NMOSD.

Optic Nerve Measurement on MRI in the Pediatric Population: Normative Values and Correlations

BACKGROUND AND PURPOSE

Few articles in the literature have looked at the diameter of the optic nerve on MR imaging, especially in children, in whom observations are subjective and no normative data exist. The aim of this study was to establish a data base for optic nerve diameter measurements on MR imaging in the pediatric population.

MATERIALS AND METHODS

This was a retrospective study on the MR imaging of pediatric subjects (younger than 18 years of age) at the Department of Diagnostic Radiology at the American University of Beirut Medical Center, Beirut, Lebanon. The optic nerve measurements were obtained by 3 raters on axial and coronal sections at 3 mm (retrobulbar) and 7 mm (intraorbital) posterior to the lamina cribrosa.

RESULTS

Of 211 scans of patients (422 optic nerves), 377 optic nerves were measured and included. Ninety-four patients were female (45%) and the median age at MR imaging was 8.6 years (interquartile range, 3.9-13.3 years). Optic nerves were divided into 5 age groups: 0-6 months (n = 18), 6 months-2 years (n = 44), 2-6 years (n = 86), 6-12 years (n = 120), and 12-18 years (n = 109). An increase in optic nerve diameter was observed with age, especially in the first 2 years of life. Measurements did not differ with eye laterality or sex.

CONCLUSIONS

We report normative values of optic nerve diameter measured on MR imaging in children from birth to 18 years of age. A rapid increase in optic nerve diameter was demonstrated during the first 2 years of life, followed by a slower increase. This was independent of sex or eye laterality.

Pattern Recognition of the Multiple Sclerosis Syndrome

During recent decades, the autoimmune disease neuromyelitis optica spectrum disorder (NMOSD), once broadly classified under the umbrella of multiple sclerosis (MS), has been extended to include autoimmune inflammatory conditions of the central nervous system (CNS), which are now diagnosable with serum serological tests. These antibody-mediated inflammatory diseases of the CNS share a clinical presentation to MS. A number of practical learning points emerge in this review, which is geared toward the pattern recognition of optic neuritis, transverse myelitis, brainstem/cerebellar and hemispheric tumefactive demyelinating lesion (TDL)-associated MS, aquaporin-4-antibody and myelin oligodendrocyte glycoprotein (MOG)-antibody NMOSD, overlap syndrome, and some yet-to-be-defined/classified demyelinating disease, all unspecifically labeled under MS syndrome. The goal of this review is to increase clinicians’ awareness of the clinical nuances of the autoimmune conditions for MS and NMSOD, and to highlight highly suggestive patterns of clinical, paraclinical or imaging presentations in order to improve differentiation. With overlay in clinical manifestations between MS and NMOSD, magnetic resonance imaging (MRI) of the brain, orbits and spinal cord, serology, and most importantly, high index of suspicion based on pattern recognition, will help lead to the final diagnosis.

Clinical characteristics and long-term visual outcome of optic neuritis in neuromyelitis optica spectrum disorder: A comparison between Thai and American-Caucasian cohorts

BACKGROUND

Neuromyelitis optica spectrum disorder (NMOSD) occurs more commonly in Asian than Caucasian populations. Few studies have examined the clinical features and visual outcome of optic neuritis (ON) within NMOSD in different racial populations. The objective of this study was to compare the clinical characteristics and long-term visual outcome of a Thai and an American-Caucasian cohort with NMOSD-related ON.

METHODS

Medical records including brain and orbital magnetic resonance imaging (MRI) of 16 consecutive subjects who developed visual loss due to ON as part of NMOSD evaluated at a single American tertiary referral center between 2006 and 2015 were reviewed and compared to those of 16 consecutive similar subjects evaluated at a single Thai tertiary referral center between 2010 and 2016. These cohorts represented the total number of NMOSD-related ON subjects seen during that time at those institutions. Statistical analyses were used for continuous and categorical data sets, and multiple regression analysis was used to adjust for differences in duration of follow-up and number of episodes of ON in each affected eye.

RESULTS

All subjects within the Thai cohort were Asian, while the American cohort initially consisted of 14 Caucasian, 1 Asian and 1 African-American subject, but the latter two were excluded from analysis. In the Thai cohort, ON occurred in 21 eyes, with a total of 19 episodes, while in the American-Caucasian cohort ON occurred in 22 eyes, with a total of 21 episodes. Aquaporin 4 (AQP4)-antibody was positive in all subjects except for one American-Caucasian subject. The mean follow-up time was 17.8 (± 16.0) and 52.8 (± 51.9) months for the Thai and American-Caucasian populations, respectively. There was no difference between the two cohorts with respect to gender, age of NMOSD and NMOSD-related ON onset, initial clinical presentation of NMOSD, initial visual acuity and automated visual fields, prevalence of swollen optic disc in the acute phase, presence of pain on the affected side, mean time of onset of ON symptoms to MRI examination, distribution of segmental involvement of the anterior visual pathway abnormalities based on MRI findings, mean time of onset of ON symptoms to treatment, final visual acuity and automated visual fields. However, a higher proportion of Thai affected eyes were found to have an initial visual acuity of 20/200 or worse compared with the American-Caucasian cohort. Azathioprine was the most common maintenance treatment (75%) used among Thai subjects in contrast to rituximab (78.6%) among American-Caucasian subjects.

CONCLUSION

Despite the different prevalence among Thai and American-Caucasian populations, the clinical characteristics of ON in the NMOSD were very similar across these two populations, other than for more severe visual loss initially among Thai subjects. Notably, long-term visual outcome did not differ between these cohorts despite significant difference in the maintenance treatment regimen. This study did not assess neurological status or outcome.