I'm Dr. Sai Kiran Maryada, a Research Scientist at the Data Institute for Societal Challenges (DISC). My work runs from deep learning for medical and subsurface imaging to full-stack dashboards used by public-health and research partners across Oklahoma.
Most research stops at the paper. I carry it through to software people sign in to — the data work, the models, and the interfaces that put results in front of decision-makers.
Integrating clinical, imaging, geospatial, and community data into pipelines that are ready for analysis.
Deep learning and image analysis for diagnosis, prognosis, and rock-property estimation.
Secure web apps with role-based access, review workflows, and controlled export.
Interactive charts and maps that turn model output into decisions.
A secure surveillance and planning dashboard that turns state and community data into accessible visualizations and decision-support views for prevention partners.
Open dashboard ↗A statewide registry of Oklahoma life-science researchers, core services, and equipment — with role-based access, review workflows, analytics, and controlled export.
Visit LSOK ↗A platform that helps OU researchers discover and track funding opportunities aligned with their expertise.
Open platform ↗An interactive dashboard exploring SNAP participation and food-assistance indicators across Oklahoma communities.
I received my Ph.D. in Computer Science and M.S. in Data Science and Analytics from the University of Oklahoma. My doctoral research introduced deep learning and multi-stage image analysis algorithms for retinal screening, breast lesion classification, ischemic stroke prognosis, and computer-aided detection — improving predictive accuracy and interpretability in real clinical research studies.
I've contributed to NSF- and NIH-funded initiatives and published in SPIE Medical Imaging, IEEE Transactions on Biomedical Engineering, and Geoenergy Science and Engineering. My work spans computer vision, predictive modeling, explainable AI, human–computer teaming, and large-scale data pipelines.
Alongside research, I design and build full-stack web platforms — React front ends, interactive Plotly, D3, and Mapbox visualizations, and secure role-based workflows — so that models leave the lab and reach the people who make decisions.
2022 Dissertation Excellence AwardOU Gallogly College of Engineering →A deep learning workflow that rapidly estimates petrophysical properties in unconventional shales directly from SEM images. Using 8,000+ images from 22 shale plays, a ResNet50-based classifier learns microstructural differences; its feature embeddings are then related to porosity, mineralogy, and elastic properties — avoiding segmentation workflows that can take weeks per image.
A fast, image-based workflow to estimate Young’s modulus from SEM images of drill cuttings across 14 shale plays. Textural and shape-based features are combined with deep learning representations and correlated with measured elastic properties via non-parametric regression, yielding strong predictive capability that holds on previously unseen images.
A self-supervised ViT-based segmentation workflow for FIB-SEM shale images that removes the need for pixel-level labels. The model segments images into 12 sub-classes grouped into organic, inorganic, and pore space, performing robustly across 22 shale plays and significantly accelerating shale characterization.
