Author Of 1 Presentation
PS16.03 - Use of machine learning classifiers based on structural and functional visual metrics to predict diagnosis in children with acquired demyelination.
Abstract
Background
Predicting diagnosis in youth at the first episode of demyelination is feasible in some but not all cases. Machine learning classifiers (MLC) can be trained to identify relationships between numerous multimodal input features and disease classifications to provide highly accurate predictions.
Objectives
To assess performance of machine learning classifiers for early disease diagnosis based on visual metrics in youth with demyelination.
Methods
Standardized clinical and visual data was prospectively collected at disease onset from 141 pediatric subjects with acquired demyelinating syndromes (ADS) and 75 healthy controls (HC). Participants were recruited through The Hospital for Sick Children (Toronto, Ontario (2010-2020)) and University of Calgary (2010-2017). Patients were classified using consensus definitions of demyelinating disorders and serum antibody testing for myelin oligodendrocyte glycoprotein (MOG) and aquaporin 4 (AQP4). Twenty-two auto-segmented Optical Coherence Tomography (OCT) features, 4 functional visual and 4 clinical features were used in a stratified manner alone or in combination to identify which combination of features provided the highest predictive accuracy. These input features were analyzed using 9 supervised MLC (Random Forest (RF), AdaBoost, XGBoost, Decision Tree (DT), Logistic Regression, Support Vector Machines (SVM), k-Nearest Neighbors, Stochastic Gradient Descent, Multinomial Naive Bayes). Data was split 80/20 between training and test sets. Backward feature selection was performed to re-run the analysis with best scoring predictor features in the MLC with highest predictive accuracy.
Results
AdaBoost, SVM, and DT were the best performing MLC with a test set accuracy between 82-88% in distinguishing between ADS and HC eyes. Multiple sclerosis (MS) was distinguished from HC with 92% accuracy. In descending order, fovea thickness, inferotemporal ganglion cell layer (GCL) thickness, low contrast visual acuity, outer inferior macular thickness, temporal peripapillary retinal nerve fiber layer and superior GCL thicknesses were the most important contributors to disease classification.
Conclusions
MLC can be used to combine visual metrics and clinical parameters to distinguish ADS from HC, and to predict MS. In addition to commonly used clinical metrics, we identified other structural and functional metrics that contribute importantly to classification. Among the machine learning algorithms tested, AdaBoost, SVM and DT performed best for this model.
Author Of 1 Presentation
P0726 - Longitudinally extensive transverse myelitis and NMOSD associated with dengue infection: a case report and systematic review of cases (ID 1155)
Abstract
Background
Neuromyelitis optica spectrum disorder (NMOSD) has been linked to para-and post-infectious triggers. Dengue virus (DENV) is one of the most common arbovirus infections in the world, with a wide spectrum of neurological clinical manifestations, including longitudinally extensive transverse myelitis (LETM), albeit rarely. It is unclear if the association is coincidental, permissive, or causal.
Objectives
We present a case of DENV-associated aquaporin-4 positive (AQP4) NMOSD with LETM with a systematic review of cases reported in the literature.
Methods
In addition to our case report, for the systematic review, we searched Medline/PubMed through April 2020, without language or design restrictions, for case reports, series and observational studies that described patients with DENV-associated LETM and/or NMOSD. We also hand-searched references and relevant conference abstracts.
Results
Case report: An adolescent girl who recently had immigrated from the Philippines presented with a 2-day history of paraparesis and urinary incontinence. Magnetic resonance imaging revealed spinal hyperintensity with patchy enhancement from T4-T7. AQP4 antibodies were positive on cell-based assay testing (Mayo Clinic Laboratories) with a titer of 1:10,000. She was diagnosed with AQP+ NMOSD. Infectious workup revealed serum +IgM and IgG antibodies against DENV, consistent with an acute dengue infection. She responded to high-dose steroids and subsequently started on rituximab maintenance. Literature review: Of 59 unique articles, 15 publications describing 18 patients met inclusion criteria. Age ranged from 8 months to 71-years (mean: 37.3) with no sex predominance. Imaging and CSF findings were heterogeneous. Four cases met 2015 criteria for NMOSD: two had LETM and optic neuritis, two had recurrent myelitis and area postrema syndrome, and one had LETM with AQP4 antibodies. The mainstay of treatment was IV methylprednisolone, although some also had IVIG and/or plasmapheresis. Prognosis varied, but few had long-term follow-up to assess for ongoing NMOSD activity.
Conclusions
DENV-associated LETM is rare, typically monophasic, and presents with severe disability, typically flaccid paraparesis. Immunotherapy should be instituted rapidly, particularly because the presentation could represent NMOSD. Decisions regarding longterm immunotherapy may depend on index of suspicion of true NMOSD, and this is where AQP4 status might be helpful. It is unknown whether there is an epidemiological or pathophysiological association between DENV infection and AQP4+ NMOSD.