Presentations
Permanent URI for this collection
Browse
Recent Submissions
Publication Deep learning-based generative network modeling for development of brain structural connectivity(2022-06) Shen, Rui Sherry; Osmanlıoğlu, Yusuf; Parker, Drew; Aunapu, Darien; Tunç, Birkan; Yerys, Benjamin E.; Verma, RaginiBrain networks are refined in the transition from childhood to adulthood. Our understanding of wiring mechanisms underlying development, however, remains very limited. Prior work has tested several predefined wiring models and suggested increased connectivity among nodes sharing similar neighbors in addition to cost minimization. While illuminating, this hypothesis-driven approach is prone to missing unforeseen network features and is biased by the researcher’s deductive thinking. Here we present a novel deep learning-based generative network model (Deep GNM) to infer developmental mechanisms in network formation in a data-informed fashionPublication Generative network models for neurodevelopment in infants with and without familial risk for autism(2022-06) Shen, Rui Sherry; Parker, Drew; Tunç, Birkan; Wang, Rongguang; Hernandez, Moises; Estes, Annette; Zwaigenbaum, Lonnie; Styner, Martin; Gerig, Guido; McKinstry, Robert; Dager, Stephen; Botteron, Kelly; Hazlett, Heather; MacIntyre, Leigh; Pandey, Juhi; Yerys, Benjamin E.; Piven, Joseph; Schultz, Robert T.; Verma, RaginiConnectivity abnormality has been widely characterized in autistic children and adolescents, but what mechanisms drive network alterations, and when network divergence arises, are unknown. Here we present a longitudinal study of structural networks over the first two years of life in 369 infants at high and low familial risk for autism. We utilize generative network models (GNMs) to explore possible wiring rules in early development and their associations with behavior scores.Publication Generative Embedding Demonstrates Altered Structural Connectivity in Autism Spectrum Disorder(2021-09) Shen, Rui Sherry; Osmanlıoğlu, Yusuf; Parker, Drew; Aunapu, Darien; Yerys, Benjamin E.; Tunç, Birkan; Verma, RaginiStudies on neurobiology of autism predominantly focus on describing topological properties of brain connectivity, offering little insight into mechanisms that give rise to associated alterations. Recently, generative network modeling has been proposed to uncover those mechanisms with parameterized wiring rules. However, most models utilize predefined wiring rules, which may not be the best characterization. Here we propose a novel deep learning based generative modeling to infer underlying mechanisms that drive atypical network growth. We demonstrate high similarity of generated network to observed structural networks, with their latent features predicting autism diagnosis. Given the neurobiological interpretability of embedding, our approach shows potential in providing connectomic biomarkers for autism.Publication Big Data, Small Bias: Harmonizing Structural Connectomes to Mitigate Site Bias in Data Integration(2025) Shen, Rui Sherry; Paker, Drew; Chen, Andrew An; Yerys, Benjamin E.; Tunç, Birkan; Roberts, Timothy P.L.; Shinohara, Russell T.; Verma, RaginiStructural connectomes are commonly used to investigate connectivity changes related to various disorders. However, small sample sizes in individual studies and highly heterogeneous disorder-related manifestations underscore the need to pool datasets across multiple studies to identify coherent and generalizable patterns linked to disorders. Yet, combining datasets introduces site bias due to variations in scanner hardware or acquisitions. This highlights the necessity for data harmonization to mitigate site bias while preserving the biological integrity associated with participant demographics and the disorders. While several paradigms exist for harmonizing normally distributed imaging data, this study represents the first effort to establish a harmonization framework specifically for structural connectomes.Publication Deriving subject-specific imaging markers for autism spectrum disorder using normative modeling on large-scale diffusion MRI data(2024-05) Shen, Rui Sherry; Parker, Drew; Wingert, Isabel A.; Reckner, Erin; Thourani, Naveen; Tunç, Birkan; Yerys, Benjamin E.; Verma, RaginiThe quest for precision medicine in autism spectrum disorder (ASD) necessitates quantitative imaging markers that account for individual variability. Traditional case-control studies offer only group-level insights, lacking subject-specific ASD information. A promising alternative is normative modeling, which defines normative developmental trajectory using large neurotypical (NT) data, and neurodevelopmental conditions are measured as individual deviations from this norm. This approach allows for the identification of personalized imaging patterns, which can be precursors to biomarkers. In this study, we applied normative modeling on large-scale diffusion MRI (dMRI) data pooled from 10 sites, aiming to 1) derive imaging markers to characterize each ASD subject, and 2) explore their relationship with autistic traits.