Your Lab Is Data Rich
Onyx Makes It Insight Rich
AI-powered microstructure analysis that processes your entire dataset — not just a spot check.
Your team acquires terabytes of EDS data
Only a fraction gets reviewed
The Problem
Modern EDS systems generate massive, multidimensional datasets. Manual review cannot keep up.
The Old Way
Spot-check 10–20 areas. Subjective. Slow.
Limited to known phases.
Miss what you don’t expect
The Onyx Way
Every pixel analyzed. Novel phases detected
automatically. Auditable, quantitative results.
Do you know how much data you need for statistical confidence and certification?
Can you repeatably analyze that much data?
More Headcount Won't Get You There
Neither Will Deep Learning
Manual review can’t scale.
Tripling your headcount moves you from 5% to 15% coverage — at significant cost. Interns need training and supervision. Senior engineers are too expensive for pixel-by-pixel review. And no amount of staffing eliminates the fundamental sampling problem: if you’re not analyzing the entire dataset, you’re gambling on what you missed
Deep learning introduces its own bottleneck.
Supervised models (CNNs) require hundreds of hours of expert labeling per material system — repeated for every alloy, every instrument setting, every magnification change. They cannot detect novel phases. And they must be retrained whenever acquisition parameters change. For many labs, this “training debt” is as prohibitive as manual review itself.
Onyx detects anomalies the first time it sees them.
Onyx is a zero-shot statistical model. No labeled training data. No retraining when parameters change. No blind spots for novel or unexpected phases. Install the software, import your EDS data, and review real, actionable insights on your first analysis — including phases that aren’t in any library. Where manual review gambles on sampling and CNNs gamble on training completeness, Onyx analyzes every pixel in every field of view, every time.
Before Onyx vs. With Onyx
|
|
Before Onyx
|
With Onyx
|
|---|---|---|
Data Management
One Library. Every Instrument. Every Collaborator.
Onyx has a built-in Laboratory Information Management System that brings all your data together in one place. It connects data from different microscopes, labs, and even other organizations. Whether your EDS data comes from your in-house Oxford system, a collaborator’s Bruker system, or outsourced work done on an EDAX system, Onyx imports and organizes everything so you can easily compare results side by side. You can search by alloy, date, instrument, or any custom detail you choose. There is no need to manage different vendor file formats or keep separate spreadsheets anymore.
Cross-Instrument Comparison
Custom Metadata Fields
Searchable Dataset Library
Multi-Lab Collaboration
Compatible With Leading Microscope Manufacturers
Our software integrates with leading microscope systems
and continues to expand compatibility across additional manufacturers.
Why Onyx
Your Data Stays Yours
Open HDF5 format. No lock-in. Free Python code included. Auditable in perpetuity.
Completely Offline — Zero IP Risk
Data never leaves your machine. No cloud dependency. Air-gap compatible. Your sensitive IP stays secure.
Built to Empower Experts
Onyx handles the drudgery so your team can focus on interpretation, trends, and decisions backed by 100% of the data.
Sparked at Oak Ridge National Laboratory
The idea behind D10 Analytics was sparked at Oak Ridge National Laboratory to address an endemic industry-wide problem: labs generate terabytes of characterization data but only analyze a fraction of it. Onyx was built to close that gap.
Our Team
Joseph Simpson
Founder
B.S. & M.S. Mechanical Engineering (Tennessee Tech), M.B.A. (University of Kansas). Former data scientist at Oak Ridge National Laboratory, focused on sparse sampling, QC of additively manufactured parts for nuclear reactors, and machine vision for electron microscopes.