Convolutional neural network-aided tuber segmentation in tuberous sclerosis complex patients correlates with electroencephalogram.

OBJECTIVE: One of the clinical hallmarks of tuberous sclerosis complex (TSC) is radiologically identified cortical tubers, which are present in most patients. Intractable epilepsy may require surgery, often involving invasive diagnostic procedures such as intracranial electroencephalography (EEG). Identifying the location of the dominant tuber responsible for generating epileptic activities is a critical issue. However, the link between cortical tubers and epileptogenesis is poorly understood. Given this, we hypothesized that tuber voxel intensity may be an indicator of the dominant epileptogenic tuber. Also, via tuber segmentation based on deep learning, we explored whether an automatic quantification of the tuber burden is feasible. METHODS: We annotated tubers from structural magnetic resonance images across 29 TSC subjects, summarized tuber statistics in eight brain lobes, and determined suspected epileptogenic lobes from the same group using EEG monitoring data. Then, logistic regression analyses were performed to demonstrate the linkage between the statistics of cortical tuber and the epileptogenic zones. Furthermore, we tested the ability of a neural network to identify and quantify tuber burden. RESULTS: Logistic regression analyses showed that the volume and count of tubers per lobe, not the mean or variance of tuber voxel intensity, were positively correlated with electrophysiological data. In 47.6% of subjects, the lobe with the largest tuber volume concurred with the epileptic brain activity. A neural network model on the test dataset showed a sensitivity of .83 for localizing individual tubers. The predicted masks from the model correlated highly with the neurologist labels, and thus may be a useful tool for determining tuber burden and searching for the epileptogenic zone. SIGNIFICANCE: We have proven the feasibility of an automatic segmentation of tubers and a derivation of tuber burden across brain lobes. Our method may provide crucial insights regarding the treatment and outcome of TSC patients.

Cross-sectional, quantitative analysis of motor function in females with HNRNPH2-related disorder.

AIMS: To describe the gross motor function of individuals with HNRNPH2-related disorder (OMIM 300986, Mental Retardation, X-linked, Syndrome, Bain Type; MRXSB) and determine the associations between clinician-measured motor function and caregiver-reported mobility scores. METHODS: Developmental histories of 17 female participants with HNRNPH2-related disorder (mean age 11.2 years, range 2.7-37.1 years) with various genotypes within and adjacent to the nuclear localization sequence (NLS) were analyzed. Participants performed the Gross Motor Function Measure-88 (GMFM-88) and caregivers completed developmental histories and the Pediatric Evaluation of Disability Inventory-Computer Adaptive Test (PEDI-CAT). RESULTS: All participants had measurable and quantifiable motor impairments. A strong positive correlation between the clinician-measured GMFM-88 total score and the caregiver-reported PEDI-CAT mobility domain score was established. Motor deficits were noted more often in individuals who were nonverbal. The 2 participants with genotypes adjacent to the NLS appear to have milder motor phenotypes. CONCLUSIONS: The GMFM-88 and PEDI-CAT are useful and feasible measures of mobility in individuals with HNRNPH2-related disorders. Convergent validity was established between the clinician-measured GMFM-88 raw scores and caregiver-reported PEDI-CAT mobility domain scores. Factors including verbal status and genotype may impact motor abilities.

Detailed Clinical and Psychological Phenotype of the X-linked HNRNPH2-Related Neurodevelopmental Disorder.

OBJECTIVE: To expand the clinical phenotype of the X-linked HNRNPH2-related neurodevelopmental disorder in 33 individuals. METHODS: Participants were diagnosed with pathogenic or likely pathogenic variants in HNRNPH2 using American College of Medical Genetics and Genomics/Association of Molecular Pathology criteria, largely identified via clinical exome sequencing. Genetic reports were reviewed. Clinical data were collected by retrospective chart review and caregiver report including standardized parent report measures. RESULTS: We expand our clinical characterization of HNRNPH2-related disorders to include 33 individuals, aged 2-38 years, both females and males, with 11 different de novo missense variants, most within the nuclear localization signal. The major features of the phenotype include developmental delay/intellectual disability, severe language impairment, motor problems, growth, and musculoskeletal disturbances. Minor features include dysmorphic features, epilepsy, neuropsychiatric diagnoses such as autism spectrum disorder, and cortical visual impairment. Although rare, we report early stroke and premature death with this condition. CONCLUSIONS: The spectrum of X-linked HNRNPH2-related disorders continues to expand as the allelic spectrum and identification of affected males increases.