Deep Learning-Based Method to Differentiate Neuromyelitis Optica Spectrum Disorder From Multiple Sclerosis.

Background: Differentiating neuromyelitis optica spectrum disorder (NMOSD) from multiple sclerosis (MS) is crucial in the field of diagnostics because, despite their similarities, the treatments for these two diseases are substantially different, and disease-modifying treatments for MS can worsen NMOSD. As brain magnetic resonance imaging (MRI) is an important tool to distinguish the two diseases, extensive research has been conducted to identify the defining characteristics of MRI images corresponding to these two diseases. However, the application of such research in clinical practice is still limited. In this study, we investigate the applicability of a deep learning-based algorithm for differentiating NMOSD from MS. Methods: In this study, we included 338 participants (213 patients with MS, 125 patients with NMOSD) who visited the Asan medical center between February 2009 and February 2020. A 3D convolutional neural network, which is a deep learning-based algorithm, was trained using fluid-attenuated inversion recovery images and clinical information of the participants. The performance of the final model in differentiating NMOSD from MS was evaluated and compared with that of two neurologists. Results: The deep learning-based model exhibited an area under the receiver operating characteristic curve of 0.82 (95% CI, 0.75-0.89). It differentiated NMOSD from MS with an accuracy of 71.1% (sensitivity = 87.8%, specificity = 61.6%), which is comparable to that exhibited by the neurologists. The intra-rater reliability of the two neurologists was moderate (κ = 0.47, 0.50), which was in contrast with the consistent classification of the deep learning-based model. Conclusion: The proposed model was verified to be capable of differentiating NMOSD from MS with accuracy comparable to that of neurologists, exhibiting the advantage of consistent classification. As a result, it can aid differential diagnosis between two important central nervous system inflammatory diseases in clinical practice.

Clinical and imaging features of patients with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy and cysteine-sparing NOTCH3 mutations.

BACKGROUND: Characteristics of patients with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) and cysteine-sparing NOTCH3 mutations are relatively unknown. This study compared clinical and imaging characteristics between patients with CADASIL and cysteine-sparing NOTCH3 mutations and those with CADASIL and cysteine-involving NOTCH3 mutations. METHODS: We retrospectively reviewed medical records of patients with CADASIL admitted to the Asan Medical Center between September 1999 and September 2017. We compared clinical and brain magnetic resonance imaging (MRI) characteristics based on the presence or absence of cysteine-involving NOTCH3 gene mutations. We compared white matter change frequencies and grades in specific spatial regions between the groups according to age-related white matter change (ARWMC) scores. We evaluated the presence, number, and anatomical distributions of cerebral microbleeds according to the microbleed anatomical rating scale. RESULTS: We reviewed data from 79 patients (55 cysteine-involving, 24 cysteine-sparing NOTCH3 mutations). Clinical symptoms and signs did not differ significantly between the groups. The white matter change frequency and ARWMC scores (adjusted for age and stroke risk factors) in the anterior temporal lobes were lower in cysteine-sparing patients than in cysteine-involving patients. Frequencies and grades of the other brain region's white matter changes and cerebral microbleeds were similar between the groups. CONCLUSIONS: Patients with CADASIL and cysteine-sparing NOTCH3 mutations showed less involvement of the anterior temporal lobes in brain MRI than those with CADASIL and cysteine-involving NOTCH3 mutations, although both groups showed similar clinical characteristics.

The clinical and prognostic value of antinuclear antibodies in NMO-IgG seropositive neuromyelitis optica spectrum disorder.

In neuromyelitis optica spectrum disorders (NMOSD), the clinical and long-term prognostic value of antinuclear antibodies (ANAs) is unclear. We analyzed registry data of NMO-IgG seropositive NMOSD patients (n = 74) according to ANA presence. The ANA-positive group (n = 32) demonstrated more frequent other autoantibodies (anti-SSA/Ro, anti-SSB/La, antiphospholipid, and anti-double stranded DNA antibodies) than did the ANA-negative group (n = 42). Clinically, annual relapse rates, and average lesion extents on MRI during attacks were comparable between the two groups (median follow-up of 7 years). The development of a poor outcome (walking with unilateral aid) also did not differ. In conclusion, although common, ANAs were not associated with a benign/malignant disease course in our NMOSD cohort.

Comparison of brain magnetic resonance imaging between myotonic dystrophy type 1 and cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy.

BACKGROUND: Anterior temporal lobe hyperintensities detected by brain MRI are a recognized imaging hallmark of cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL). Because similar findings may be present in patients with myotonic dystrophy type 1 (DM1), the brain MRI in these two diseases is often misinterpreted. We compared the MRI findings between the two entities to examine whether they display distinctive characteristics. METHODS: This retrospective, cross-sectional study reviewed medical records of patients with DM1 or CADASIL admitted to Asan Medical Center between September 1999 and September 2017. We compared the frequency and grades of white matter changes in specific spatial regions between the groups according to age-related white matter change scores. We also evaluated the presence of cerebral microbleeds. RESULTS: A total of 29 patients with DM1 and 68 with CADASIL who had undergone MRI were included in the analysis. The overall prevalence of white matter hyperintensities was 20 (69%) and 66 (97%) in DM1 and CADASIL, respectively (p < 0.001), whereas the frequency of anterior temporal lobe hyperintensities was comparable between the groups (10 [34.5%] in DM1 vs. 35 [51.5%] in CADASIL, p = 0.125). The brain MRI of patients with DM1 revealed more limited involvement of the frontal, parieto-occipital, external capsule and basal ganglia regions compared with imaging in patients with CADASIL. Cerebral microbleeds were not observed in any case of DM1 but were present in 31 of 45 (68.9%) cases of CADASIL. CONCLUSIONS: Anterior temporal lobe involvement in DM1 is not infrequent compared with CADASIL. However, because brain MRI in patients with DM1 lacks other distinctive features seen in CADASIL, imaging might assist in differentiating these two conditions.

Factors predicting remission in thymectomized patients with acetylcholine receptor antibody-positive myasthenia gravis.

INTRODUCTION: Although thymectomy is an important therapeutic option for myasthenia gravis (MG), factors predicting remission after thymectomy are not well known. METHODS: We retrospectively reviewed patients with acetylcholine receptor antibody (AChR-Ab)-positive MG who had undergone thymectomy. Prognostic factors predicting remission were investigated. Changes in AChR-Ab titer before and after thymectomy were also evaluated. RESULTS: Among the 179 patients, 52.5% achieved complete stable or pharmacologic remission. Nonthymomatous pathology and mild preoperative status were favorable predictors of remission. The decrease in AChR-Ab titer after thymectomy was significant in nonthymomatous MG but not in thymomatous MG. DISCUSSION: Nonthymomatous pathology and mild preoperative status are prognostic factors that may predict remission after thymectomy. The decrease in AChR-Ab titer after thymectomy was significant in nonthymomatous MG but not in thymomatous MG, suggesting that the pathogenic role of the thymus differs according to pathology. Muscle Nerve 58:796-800, 2018.

Correlation between the CAG repeat size and electrophysiological findings in patients with spinal and bulbar muscular atrophy.

INTRODUCTION: Spinal and bulbar muscular atrophy (SBMA) is caused by the expansion of a CAG repeat in the androgen receptor gene. The relationship between the CAG repeat size and electrophysiological findings is not completely understood. METHODS: We retrospectively analyzed 62 SBMA patients to assess the correlation between their CAG repeat size and electrophysiological findings. RESULTS: In multiple regression analysis including age at examination and disease duration, we identified a negative correlation between the CAG repeat size and the compound muscle action potential (CMAP) amplitude. No significant correlation was found between the CAG repeat size and sensory nerve action potential (SNAP) amplitude. DISCUSSION: Contrary to previous reports of motor- and sensory-dominant phenotypes correlating with CAG repeat sizes, the CAG repeat size was negatively correlated only with CMAP amplitude, and not with SNAP amplitude. Muscle Nerve 57: 683-686, 2018.

Prevalence and Characteristics of Unruptured Cerebral Aneurysms in Ischemic Stroke Patients.

BACKGROUND AND PURPOSE: The prevalence of unruptured cerebral aneurysms (UCAs) in ischemic stroke patients is not clearly defined. This study aimed to measure the prevalence and characteristics of UCAs in patients with acute ischemic stroke (AIS) and to compare our findings with those of the general population. In addition, we investigated the factors associated with cerebral aneurysms in AIS patients. METHODS: We retrospectively reviewed the brain magnetic resonance angiography images of 955 patients with AIS and 2,118 controls who had received a brain magnetic resonance angiography as part of a health check-up. We investigated the prevalence, size, location, and risk factors of the subjects in the context of UCAs. RESULTS: UCAs were found in 74 patients with AIS (7.7%) and in 79 who received a health check-up (3.7%). The prevalence of UCAs was significantly higher in the AIS group than in the health check-up group (odds ratio 2.17, 95% confidence interval 1.56-3.01). The mean aneurysm diameter was larger in the AIS group than in the health check-up group (3.75 mm vs. 3.02 mm, P=0.009). UCAs were primarily located in the internal carotid artery in both groups, and aneurysms in the middle cerebral artery were particularly common in the AIS group. According to multivariate analysis, hypertension alone was associated with an increased prevalence of UCAs in stroke patients. CONCLUSIONS: This study identified a higher prevalence and larger size of UCAs in AIS patients than in the general population. Hypertension was an independent risk factor of UCA in AIS.

Subdural hemorrhage mimicking peripheral neuropathy.

Subdural hemorrhage (SDH) can manifest various neurologic symptoms. However, SDH presenting with only hand weakness has rarely been reported. We report two SDH cases with only hand weakness mimicking peripheral neuropathy. Since SDH can present with hand weakness only, we suggest the clinicians to do a careful history taking and recommend a CT scan in the elderly patients.