Montgomery High School

Advances in remote sensing with higher-resolution images provide new opportunities for monitoring tasks in forestry, ecology, agriculture, energy, etc. However, applications at large scale often face the challenge of spatial variability, where patterns vary over geographic regions. Existing methods often assume a fixed, known spatial partitioning as input, which is mostly unavailable in practice, or exhaust models over locations that are computationally prohibitive for large-scale Earth monitoring. We propose an adaptive router approach, which leverages large, pretrained multimodal foundation models as a model router to automatically select scenario-specialized segmentation models for each input image. We evaluate our method using high-resolution tree mortality mapping, which has been recognized as a critical gap in large-scale carbon monitoring. The new method shows over 30 % improvement in polygon F1 score over baselines including vision foundation models.

Objective: Our project aims to use Artificial Intelligent (AI)-based methods to address overlooked issues in adolescent nutrition/diet and microbiome, and to identify biomarkers and risk factors that affect long-term health outcomes. Hypothesis: Many diseases, including diabetes, dementia, and Alzheimer’s disease, are linked to microbiome dysbiosis and diet-related risk factors, which can develop during adolescence and influence health outcomes throughout adulthood. Methods and Data: We conducted a systematic literature review on observational, clinical, and laboratory data to examine the relationships among diet quality, nutrition component, microbiome profiles, and energy expenditure, and their effects on behavior and health outcomes/diseases across adolescence, adulthood, and elderly populations (65+ yrs). We designed an analysis workflow to systematically extract and categorize data. Using AI-based methods we constructed knowledge models to uncover the relationships among key concepts on nutrition, diet, microbiome, genetic factors, age, sex, socioeconomic factors, and disease outcomes. Using the citations from PubMed literature, the Dietary Guidelines for Americans, and recommendations from authoritative health organizations, we developed a novel ChatGPT-based chatbot platform, Teen Fuel Up, designed to provide teens with real-time assistance to published knowledge about diet and nutrition. Conclusions: AI-based methods are highly effective in knowledge discovery and health promotion, identifying connections between adolescent nutrition, the microbiome, and chronic diseases later in life, as well as their differences using sex as a biological variable to increase scientific understanding of conditions that are specific to women. We will continue developing the ChatGPT-based chatbot to provide real-time cited knowledge that adapts to the evolving dietary needs of individuals. National Institute of Allergy And Infectious Diseases of the National Institutes of Health, grant number R01AI130460 and U24AI171008, and CPRIT award RP220244. This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.

We present a high-accuracy stereo-vision-based dataset and a generalized image-to-Z framework for 3D reconstruction from a single 2D image, addressing data scarcity and improving prediction accuracy over state-of-the-art UNet models.

Keratinocytes are an essential component of the epidermis that undergoes constant proliferation and differentiation. However, the dysregulation of keratinocyte differentiation has been implicated in various skin disorders such as psoriasis. Imiquimod, otherwise known as IMQ, is a topical immunomodulator often used to induce psoriasis-like lesions in murine models for research purposes. This study focuses on the efficacy of using IMQ to induce a psoriasis-like model on murine skin cells by analyzing single-cell RNA sequencing and trajectory analysis. The results indicate a few differences between IMQ-induced and control murine cells, primarily the increased keratinocyte and immune cell populations, which reflects the cell identity found on psoriatic skin. However, trajectory analysis reveals that IMQ-induced cells have quite a linear differentiation pattern compared to the branched pattern found in control cells. As a result, further research must be conducted to explore differing factors between psoriatic cells and IMQ-induced cells to determine its usefulness in mimicking psoriasis-like conditions for research.

The hippocampus is a crucial brain structure involved in memory formation, spatial navigation, emotional regulation, and learning. An accurate MRI image segmentation of the human hippocampus plays an important role in multiple neuro-imaging research and clinical practice, such as diagnosing neurological diseases and guiding surgical interventions. While most hippocampus segmentation studies focus on using T1-weighted or T2-weighted MRI scans, we explore the use of diffusion-weighted MRI (dMRI), which offers unique insights into the microstructural properties of the hippocampus. Particularly, we utilize various anisotropy measures derived from diffusion MRI (dMRI), including fractional anisotropy, mean diffusivity, axial diffusivity, and radial diffusivity, for a multi-contrast deep learning approach to hippocampus segmentation. To exploit the unique benefits offered by various contrasts in dMRI images for accurate hippocampus segmentation, we introduce an innovative multimodal deep learning architecture integrating cross-attention mechanisms. Our proposed framework comprises a multi-head encoder designed to transform each contrast of dMRI images into distinct latent spaces, generating separate image feature maps. Subsequently, we employ a gated cross-attention unit following the encoder, which facilitates the creation of attention maps between every pair of image contrasts. These attention maps serve to enrich the feature maps, thereby enhancing their effectiveness for the segmentation task. In the final stage, a decoder is employed to produce segmentation predictions utilizing the attention-enhanced feature maps. The experimental outcomes demonstrate the efficacy of our framework in hippocampus segmentation and highlight the benefits of using multi-contrast images over single-contrast images in diffusion MRI image segmentation.

The germinal center response or reaction (GCR) is a hallmark event of adaptive humoral immunity. Unfolding in the B cell follicles of the secondary lymph organs, a GC culminates in the production of high-affinity antibody-secreting plasma cells along with memory B cells. By interacting with follicular dendritic cells (FDC) and T follicular helper (Tfh) cells, GC B cells exhibit complex spatiotemporal dynamics. Driving the B cell dynamics are the intracellular signal transduction and gene regulatory network that responds to cell surface signaling molecules, cytokines, and chemokines. As our knowledge of the GC continues to expand in depth and in scope, mathematical modeling has become an important tool to help disentangle the intricacy of the GCR and inform novel mechanistic and clinical insights. While the GC has been modeled at different granularities, a multiscale spatial simulation framework - integrating molecular, cellular, and tissue-level responses - is still rare. Here, we report our recent progress toward this end with a hybrid stochastic GC framework developed on the Cellular Potts Model-based CompuCell3D platform. Tellurium is used to simulate the B cell intracellular molecular network comprising NF-κB, FOXO1, MYC, AP4, CXCR4, and BLIMP1 that responds to B cell receptor (BCR) and CD40-mediated signaling. The molecular outputs of the network drive the spatiotemporal behaviors of B cells, including cyclic migration between the dark zone (DZ) and light zone (LZ) via chemotaxis; clonal proliferative bursts, somatic hypermutation, and DNA damage-induced apoptosis in the DZ; and positive selection, apoptosis via a death timer, and emergence of plasma cells in the LZ. Our simulations are able to recapitulate key molecular, cellular, and morphological GC events including B cell population growth, affinity maturation, and clonal dominance. This novel modeling framework provides an open-source, customizable, and multiscale virtual GC simulation platform that enables qualitative and quantitative in silico investigations of a range of mechanic and applied research questions in future.

This Indigenous-led project aims to better understand historical and contemporary ways in which Alaska Native Peoples steward salmon and the values connected to salmon stewardship. Indigenous Peoples have stewarded Alaska lands and waters for thousands of years yet have been largely excluded from western science and management systems. In this project, we utilize a participatory approach that allows for the equitable valuing of ideas and knowledges to document the breadth and depth of Yup'ik and Athabascan knowledge and governance systems in southwestern Alaska. We reshape research methodologies by centering Indigenous frameworks and methodologies, including circle dialogues and multi-generational interviews led by Indigenous scholars and students in their home communities and regions. In this paper, we share the Yup'ik and Athabascan values, knowledge, management, and governance mechanisms that can improve the long-term sustainability and equity of Alaska salmon systems. This research elevates the voices of Alaska Native salmon stewards and experts from the Kuskokwim Bay and the Kuskokwim River. We elaborate on five key themes that emerged from this research, including traditional Indigenous ways of life, Indigenous stewardship, self-determination, food and livelihood sovereignty, and ecosystem changes, and identify a more equitable and sustainable path forward for salmon and people in Alaska.

Bamboo composite is an attractive candidate for structural materials in applications such as construction, the automotive industry, and logistics. However, its development has been hindered due to the use of harmful petroleum-derived synthetic adhesives or low-bonding biobased adhesives. Herein, we report a novel bioadhesion strategy based on in situ lignin bonding that can process natural bamboo into a scalable and high-performance composite. In this process, lignin bonds the cellulose fibrils into a strong network via a superstrong adhesive interface formed by hydrogen bonding and nanoscale entanglement. The resulting in situ glued-bamboo (glubam) composite exhibits a record-high shear strength of ∼4.4 MPa and a tensile strength of ∼300 MPa. This in situ lignin adhesion strategy is facile, highly scalable, and cost-effective, suggesting a promising route for fabricating strong and sustainable structural bamboo composites that sequester carbon and reduce our dependence on petrochemical-based adhesives.