Better Brain Biomarkers

Radiata develops brain biomarkers and therapies to measure and improve brain health. We apply human neuroscience, AI + ML, mathematical modeling, and clinical science to promote optimal brain function in health and disease. CogFi is our first validated scan: a 30-minute brain function and structure assessment to measure neuro-cognitive health, estimate how cognitive fitness may change, and track changes over time.

Brain activity

Science

We focus on brain function biomarkers using resting state functional MRI. FMRI can precisely measure brain network function while a subject lays comfortably in an MRI scanner. Healthy brain function is the foundation for cognitive and mental well-being.

Functional and structural MRI analysis

Modern pipelines to process scans and quantify brain activity, connectivity, and anatomy.

Systems neuroscience

Network and gradient methods for revealing how distributed brain systems are organized.

Aging & dementia spectrum

Tests and therapies for Alzheimer’s disease, frontotemporal dementia, Mild Cognitive Impairment, Subjective Cognitive Concerns, and healthy aging trajectories.

Machine learning biomarkers

Deep learning models and dimensionality reduction for identifying data-driven patterns and creating latent spaces. Best practices for validation and interpretability.

Public + private datasets

Processed and cataloged public datasets (ADNI, OASIS, etc.) combined with private datasets enhance biomarker discovery and validation.

Mathematical modeling

Dynamical systems and control theory techniques for identifying brain dysfunction and developing optimal therapies.

Advanced statistics

Mixed-effects and Gaussian additive models for longitudinal and multi-site inference.

Neuropsychology

Brain-cognition relationships for quantifying cognitive processes like memory, attention, executive function, and language performance.

Technology

We are a full‑stack brain imaging biomarker platform: brain scanning, image processing, data science/neuroscience/AI, R&D, and brain health reporting.

Modern neuroimaging workflows

fMRIPrep, XCP‑D, CAT12, FSL, SPM, and AFNI; inputs and outputs adhere to BIDS standards for interoperability.

Modern scientific computing

Python, MATLAB, and R.

Data science & ML toolchain

NumPy/SciPy/pandas, scikit‑learn, PyTorch, and Hugging Face support unsupervised and supervised modeling and data/model transparency.

AWS backbone

S3 for data, RDS (PostgreSQL) for databases, EC2 for CPU/GPU compute, Batch and Step Functions for scalable processing, Lambda for automated tasks, ECS + ECR for container hosting, and IAM for security and access control.

Containerized workflows

All services and pipelines are packaged in Docker containers and stored on ECR for reproducible processing.

Interactive visualization dashboard

All data is centralized in our Brains Browser, built on React, D3.js, Tailwind CSS, and Niivue. Personalized brain health reports are built on this foundation.

DevOps & observability

Git for version control, GitHub Actions for CI/CD, and CloudWatch for logging and metrics keep the platform reliable and audit‑ready.

Scanner Integration

Automatically push DICOM files from the scanner to AWS S3 for seamless end-to-end data handling.

Team

Founded by UCSF neuroscientists and data engineers, Radiata emerged from the Memory and Aging Center with a goal to support better brain and mind health.

Core

  • Jesse Brown, PhD

    Jesse Brown, PhD

    CEO & Founder

    jesse@radiata.ai

  • Clayton Young, BS

    Clayton Young, BS

    CTO & Cofounder

    clayton@radiata.ai

Advisors

  • Ioannis Pappas, PhD

    Ioannis Pappas, PhD

    Scientific Advisor

Publications

Functional network collapse in neurodegenerative disease

Jesse A. Brown, Alex J. Lee, Kristen Fernhoff, Taylor Pistone, Lorenzo Pasquini, Amy B. Wise, Adam M. Staffaroni, Maria Luisa Mandelli, Suzee E. Lee, Adam L. Boxer, Katherine P. Rankin, Gil D. Rabinovici, Maria Luisa Gorno Tempini, Howard J. Rosen, Joel H. Kramer, Bruce L. Miller, William W. Seeley, ADNI

Nature Communications (2025)

A dynamic gradient architecture generates brain activity states

Jesse A. Brown, Alex J. Lee, Lorenzo Pasquini, William W. Seeley

NeuroImage (2022)

Regional Aβ‑tau interactions promote onset and acceleration of Alzheimer’s disease tau spreading

W. J. Lee, J. A. Brown, H. R. Kim, R. La Joie, H. Cho, C. H. Lyoo, et al.

Neuron (2022)

Patient‑Tailored, Connectivity‑Based Forecasts of Spreading Brain Atrophy

J. A. Brown, J. Deng, J. Neuhaus, I. J. Sible, A. C. Sias, S. E. Lee, et al.

Neuron (2019)

Longitudinal default mode connectivity varies with age and tracks episodic memory & processing speed

A. M. Staffaroni, J. A. Brown, K. B. Casaletto, F. M. Elahi, J. Deng, et al.

Journal of Neuroscience (2018)

Clinicopathological correlations in behavioral variant FTD

D. C. Perry, J. A. Brown, K. L. Possin, S. Datta, A. Trujillo, et al.

Brain (2017)

Advancing functional dysconnectivity and atrophy in progressive supranuclear palsy

J. A. Brown, A. Y. Hua, A. Trujllo, S. Attygalle, R. J. Binney, et al.

NeuroImage: Clinical (2017)

The UCLA Multimodal Connectivity Database: a web platform for connectivity matrix sharing & network analysis

J. A. Brown, J. D. Rudie, A. Bandrowski, J. D. Van Horn, S. Y. Bookheimer

Frontiers in Neuroinformatics (2012)

Brain network local interconnectivity loss in aging APOE‑4 allele carriers

J. A. Brown, K. H. Terashima, A. C. Burggren, L. M. Ercoli, K. J. Miller, et al.

PNAS (2011)

Partners

We work with imaging networks, neurotech companies, pharma, CROs, academic labs, and anyone interested in bringing modern brain MRI biomarkers and therapies into real-world practice and clinical trials. Interested in partnering? Contact us.


Our Partners and Supporters

  • Berkeley SkyDeck
  • NVIDIA Inception Program