Laboratory Intelligence. Engineered.

Transform microscopy data and analytical expertise into automated, reproducible, and reusable workflows.

MATERIA unifies microscopy data, AI development, analytical automation, and laboratory expertise in one vendor-neutral, discipline-agnostic software environment.

One infrastructure. Different laboratories. Their own methods.

Explore the PlatformSee how it works ↓
Under active development · Pilot collaborations welcome
Post-acquisition infrastructure for laboratories
MATERIAProjects / coatings-batch-A / Sample 07 — cross-section
ILLUSTRATIVE DEMO DATA
✓ Ingest✓ Pre-process✓ Segment✓ MeasureReviewExport
t1t2t3t4t5#3 ✓#5 ✓#8 ✓#7 ?MOUNTSUBSTRATE50 µm
CoatingPoresInclusionsCrackSchematic cross-section · not a measured sample
The problem

Advanced instruments. Fragmented analysis.

Microscopes have become highly capable. The analysis after acquisition often has not: data passes through exports, personal scripts, manual measurements and spreadsheets, and the method itself is rarely written down. The core problem is not slow segmentation. It is the absence of connected analytical infrastructure.

  1. 01Repetitive manual analytical workSpecialists repeat the same measurements across samples and batches.
  2. 02Disconnected tools and proceduresEach step happens in a different program, with manual hand-offs.
  3. 03Difficult reproducibility and traceabilityParameters and decisions are rarely stored with the result.
  4. 04Expertise locked in individualsMethods live in scripts and habits that are hard to share or hand over.
TODAYFragmented laboratory work
Microscopy acquisitioninstrument PC
manual export
File exportsshared drive
copied to a laptop
Individual scripts & softwarepersonal laptop
parameters undocumented
Manual measurementsby hand
values re-typed
Spreadsheetsspreadsheet
copy & paste
Reportsdocument

Each hand-off loses context. The result is hard to reproduce or hand over.

WITH MATERIAConnected laboratory infrastructure
MATERIA
Laboratory datasetsstructured
Versioned analytical methodsv4.0.2
Automated workflowsexecuted
Reviewed resultsapproved
Reproducible reportstraceable

One history from dataset to report, inside one structured platform.

The platform

One infrastructure. Every analytical capability.

From datasets and AI models to analytical procedures, projects, and results, MATERIA brings a laboratory’s post-acquisition analytical work into one structured environment — where methods can be versioned, improved, reproduced, and automated.

01
FOUNDATION

Laboratory Infrastructure

The structured foundation where a laboratory’s analytical assets are organized, managed, versioned, preserved and reused.

  • Centralized datasets and projects
  • Model and method registries
  • Version histories
  • Organization-managed resources
  • Traceable results
  • Collaboration and shared expertise
develops
02
DEVELOPED ON THE FOUNDATION

Laboratory Intelligence

The computational intelligence and expertise the laboratory develops and maintains itself.

  • AI and computational models
  • Laboratory-specific methods
  • Validation and improvement
  • Expert feedback
  • Reusable institutional knowledge
is executed through
03
EXECUTES THESE CAPABILITIES

Analytical Automation

Complete analytical procedures run as reusable workflows, combining models, measurements and specialist review.

  • Configurable workflows
  • Automated processing
  • Quantitative measurements
  • Batch execution
  • Specialist intervention
  • Structured reporting
DISCIPLINE-AGNOSTIC BY DESIGNThe analytical infrastructure remains consistent. The methods, models, and workflows adapt to the laboratory’s scientific domain.

Built around reusable computational components rather than discipline-specific procedures, MATERIA adapts to different laboratory environments without changing its underlying infrastructure.

GitHub for laboratory intelligence. Built for microscopy.

A shared home for your laboratory’s data, AI models, analytical methods, and workflows — where everything can be versioned, reused, executed, and continuously improved.

DevelopVersionExecuteImprove
A new kind of laboratory infrastructure

Your laboratory’s analytical work. In one place.

One organization-managed place to organize data, manage models, version procedures, collaborate across projects, and turn established methods into automated workflows.

Imaging Laboratoryorganization-managed resourcesILLUSTRATIVE
ASSETS
Coating QCVERSION HISTORY
  1. Published for the QC teamMethod owner4.0.2
  2. Model updated to coat-seg 1.3.0Method owner4.0.0
  3. Porosity statistics addedResearcher3.2.0
  4. First automated versionMethod owner3.0.0
EXECUTE
WorkflowCoating QC 4.0.2
Datasetcoatings-batch-A v3
  1. Ingest—
  2. Pre-process—
  3. Segment—
  4. Measure—
  5. Review—
  6. Export—
Each run is recorded in the project history with the exact versions used.
REPOSITORYDatasets, models, methods and results live in one organization-managed place.
VERSION HISTORYEvery change to a model or procedure is recorded and can be compared.
COLLABORATIONTeams work on shared methods instead of private copies.
+ EXECUTIONMethods do not just sit in storage. They run as workflows and produce results.
How it works · Available in alpha

From microscopy data to reproducible results.

Repetitive operations run automatically. Specialists can inspect, adjust and correct results at any stage; the dedicated Review stage is the consolidated final review and approval.

  1. 01IngestImport, organize, and structure microscopy datasets.AUTOMATED→ Structured datasetInspect datasets, metadata, and grouping.
  2. 02Pre-processApply corrections and standardize analytical inputs.CONFIGURABLE→ Standardized imagesAdjust settings and compare results.
  3. 03SegmentIdentify structures, phases, and objects using AI or computational methods.AI-ASSISTED · CONFIGURABLE→ Masks and objectsCorrect masks, boundaries, and classifications.
  4. 04MeasureQuantify properties and generate statistics.AUTOMATED→ Measurement tablesFilter objects and verify statistics.
  5. 05ReviewConsolidate results and approve the analysis.FINAL REVIEW→ Approved resultsResolve open issues and approve.
  6. 06ExportGenerate reports, datasets, and traceable results.AUTOMATED · CONFIGURABLE→ Report + provenanceCheck report configuration and outputs.
◆ SPECIALIST INSPECTION & CORRECTION — AVAILABLE THROUGHOUT THE WORKFLOWAUTOMATEDCONFIGURABLE◆ OPTIONAL INTERVENTIONCONSOLIDATED FINAL REVIEW
Early laboratory results — FCH VUT

From over an hour of analysis to minutes.

In early coatings quality-control work at FCH VUT, an analytical procedure previously requiring approximately 60–90 minutes was reported to take around 5–10 minutes using MATERIA-assisted automation.

BEFORE60–90minConventional analytical procedure
WITH MATERIA5–10minMATERIA-assisted workflow
TIME REDUCTION≈83–94%less analysis time, depending on the comparison
ELAPSED TIME PER ANALYSIS · MINUTES
Conventional
With MATERIA
0306090
Hatched segments show the reported range.

Early application-specific result. Timing and analytical scope are subject to further validation. Not a general performance benchmark.

Workflow BuilderIN DEVELOPMENT

Build a method once. Reuse it across projects.

Laboratories will be able to compose their own procedures from reusable components with typed interfaces and configurable parameters, and publish them as versioned workflows. The modular architecture is partly in place; the graphical builder shown here is an illustrative preview.

Coating QCv4.0.2 · publishedv4.1.0 · draftillustrative preview · not generally available
Load datasetio.load 1.0→ image[]Normalizenormalize.hist 1.3image[] → image[]Denoisedenoise.nlm 1.2image[] → image[]Segmentsegment.apply 2.0image[] → mask[]Measure thicknessmeasure.thickness 1.1mask[] → tablePorosity & defectsstats.porosity 1.0mask[] → tableReportreport.technical 2.0table → pdf, csvMODEL · coat-segv1.3.0 · lab-trained

Select a component to inspect its parameters.

Architecture

Modular by design. Reproducible by default.

Reusable computational building blocks compose into complete analytical workflows. Versioned AI models can be integrated wherever needed.

LEVEL 1 — COMPUTATIONAL COMPOSITION
BUILDING BLOCKFunctionsProcessing, transformation, inference and measurement operations.
used by
COMPONENTFragmentsReusable components with defined interfaces and parameters.
coordinated by
COMPOSITIONProcessesCoordinate Fragments to accomplish a larger analytical task.
composed into
PROCEDUREWorkflowsComplete procedures: sequence, configuration, execution logic.
PARALLEL RESOURCE · VERSIONEDModelsSeparately managed, versioned AI models, including the laboratory’s own. Used where needed; AI is optional.
pinned version
LEVEL 2 — EXECUTION & RESULTSA Project is the execution environment; it uses a pinned Workflow version.
VERSIONED PROCEDUREPinned WorkflowA specific version of a reusable analytical procedure.
selected by
EXECUTION ENVIRONMENTProject ExecutionExecutes the pinned procedure against laboratory datasets.
produces
RESULTSTraceable OutputsMeasurements, reports, datasets and results linked to the methods, models, parameters and versions that produced them.
LEVEL 1 — COMPUTATIONAL COMPOSITION
BUILDING BLOCKFunctionsProcessing, transformation, inference and measurement operations.
↓ used by
COMPONENTFragmentsReusable components with defined interfaces and parameters.
↓ coordinated by
COMPOSITIONProcessesCoordinate Fragments to accomplish a larger analytical task.
↓ composed into
PROCEDUREWorkflowsComplete procedures: sequence, configuration, execution logic.
┆ alongside, used where needed
PARALLEL RESOURCE · VERSIONEDModelsSeparately managed, versioned AI models, including the laboratory’s own. Used where needed; AI is optional.
LEVEL 2 — EXECUTION & RESULTS
VERSIONED PROCEDUREPinned WorkflowA specific version of a reusable analytical procedure.
↓ selected by
EXECUTION ENVIRONMENTProject ExecutionExecutes the pinned procedure against laboratory datasets.
↓ produces
RESULTSTraceable OutputsMeasurements, reports, datasets and results linked to the methods, models, parameters and versions that produced them.
VersioningEvery procedure keeps its configuration and version, so results can be reproduced, reviewed and compared.
ProvenanceEach output records the data, parameters, model and workflow versions behind it.
ReproducibilityA project can re-run a past analysis with the same pinned versions.
ConfigurabilityParameters are explicit and documented instead of hidden inside scripts.
Organization-level managementTeams see which methods exist, who maintains them, and where they are used.
Reuse across projectsA validated method is applied to new datasets without being rebuilt.
BRING YOUR OWN MODELSIN DEVELOPMENT

Your models. Your methods. Your analytical infrastructure.

MATERIA is designed to integrate laboratory-developed and externally developed AI models through defined interfaces, so laboratories are not limited to models supplied by the platform. Laboratories can also develop and reuse their own Functions, Fragments, Processes and Workflows.

Laboratories will be able to bring their own computational methods and compatible AI models as these integration capabilities are implemented.

The Annotator · AI data preparationIN DEVELOPMENT

Better training data. Less annotation work.

An optimized workspace for creating, correcting, and managing microscopy segmentation masks. The Annotator helps laboratory specialists transform raw images into structured, high-quality datasets for training their own AI models.

Designed around efficient annotation and expert control, The Annotator connects data preparation directly to MATERIA’s model-development and analytical workflow ecosystem.

The Annotatorcoatings-training-set / image 014ILLUSTRATIVE INTERFACE
ORIGINAL IMAGEread-only
Original microscopy image
SEGMENTATION MASKeditable
Editable segmentation mask
Efficient mask creationCreate and refine segmentation masks with a streamlined, purpose-built interface.
Specialist-controlled annotationInspect, correct, and validate object boundaries and class assignments.
Structured training datasetsOrganize annotated microscopy images and their corresponding masks for model development.
Connected model improvementPrepare reviewed annotations for reuse in training and improving laboratory-specific AI models.
WHERE THE ANNOTATOR FITS
  1. Microscopy images
  2. →The Annotator
  3. →Labeled datasets
  4. →Model training
  5. →Validated AI models
  6. →Analytical workflows
↓ Training, validation and improvement continue in Laboratory Intelligence
Laboratory Intelligence

Your laboratory’s expertise, preserved and continuously improved.

Laboratory Intelligence is more than training models. It is the systematic conversion of specialist knowledge into structured assets the laboratory owns: models that improve with review, and methods that stay reproducible after people move on.

Model intelligenceIn development

Laboratories develop, validate, deploy and improve their own computational models.

  1. 01→AnnotateSpecialists label structures in their own images.
  2. 02→TrainLaboratory-specific models are trained on that data.
  3. 03→ValidateVersions are compared against reference sets.
  4. 04→DeployValidated models are used inside workflows.
  5. 05→ReviewResults are inspected and corrected by experts.
  6. 06↻ImproveCorrections inform the next version.
Method intelligencePartly in alpha

Laboratories develop, version, validate, share and reuse complete analytical procedures.

  1. Researcher knowledgetacit
  2. Configured analytical methodexplicit parameters
  3. Validated workflowchecked on references
  4. Versioned laboratory assetorganization-managed
  5. Reuse across projectspinned versions
  6. Improvement from new resultsnext version
Expert inspection and correction at every stageAvailable in alpha
Traceable analytical decisionsAvailable in alpha
Reusable analytical proceduresIn development
Validated configurations stored with each methodIn development
Organization-level sharing of validated methodsIn development
Controlled development of laboratory-specific modelsIn development
Where MATERIA fits

Independent of instruments. Adaptable to disciplines.

MATERIA operates above the acquisition layer, providing a common environment for analytical data, computational methods, AI models, and reproducible workflows. Laboratories retain the freedom to develop and use the methods appropriate to their scientific work.

INSTRUMENTS, ANY VENDOR
SEMelectron
AFM / SPMprobe
Optical microscopylight
TEMelectron
Archived datasetsfiles
data
ONE ANALYTICAL ENVIRONMENT
MATERIAAnalytical automationLaboratory intelligenceLaboratory infrastructure
OUTCOMES
Research outcomesReproducible quantitative results, preserved methods
Industrial outcomesConsistent evaluations, structured reports
INSTRUMENTS, ANY VENDOR
SEMelectron
AFM / SPMprobe
Optical microscopylight
TEMelectron
Archived datasetsfiles
↓ microscopy data
MATERIAAnalytical automationLaboratory intelligenceLaboratory infrastructure
↓
Research outcomesReproducible quantitative results, preserved methods
Industrial outcomesConsistent evaluations, structured reports

MATERIA is designed to be vendor-neutral. Instrument integrations are developed and validated progressively; support depends on the instrument and data format.

Research laboratories

  • —Reduce repetitive analytical work
  • —Make complex methods reproducible
  • —Reuse validated methods across studies
  • —Support collaboration between researchers
  • —Depend less on individual scripts and specialists

Industrial and quality-control laboratories

  • —Standardize recurring evaluations
  • —Improve analytical consistency
  • —Process larger datasets efficiently
  • —Generate structured reports
  • —Maintain traceable analytical procedures

Microscopy manufacturers

  • —Extend the analytical value of instrument data
  • —Explore advanced post-acquisition workflows
  • —Offer customers specialized analytical capabilities
  • —Investigate integrations without building every method in-house
  • —Explore complementary software partnerships

Potential benefits. Outcomes depend on the application and are evaluated in pilots.

Applications · Pilots & Research Collaborations

Built around real laboratory challenges.

Current development applications illustrate the platform’s capabilities, not the limits of its intended use.

Coatings QC microscopy image
FCH VUTFaculty of Chemistry, Brno University of Technology

Coatings Quality Control

Development & Validation
The challenge

Manual evaluation of coating thickness, porosity, cracks, inclusions, and interface quality is repetitive and time-consuming.

The approach

Reusable analytical workflows combine image processing, segmentation, quantitative measurement, and specialist review.

The outcome

Consistent, reproducible analysis and structured reports.

60–90 min → 5–10 minEarly resultView Case Study →
Advanced microscopy image
NenoVisionAdvanced microscopy

Advanced Microscopy

Pilot — In Development
The challenge

Advanced microscopy generates complex datasets requiring specialized post-acquisition analysis.

The approach

Exploring reusable analytical workflows and potential integration with advanced microscopy systems.

The outcome

Evaluating additional analytical capabilities for microscopy users.

PilotEvaluating workflow fit and potential integration.
Materials research image
ÚMVI FSI VUTInstitute of Materials Science and Engineering

Materials Research

Research Engagement — In Development
The challenge

Research methods often depend on individually maintained scripts and procedures, limiting reproducibility and reuse.

The approach

Exploring versioned analytical methods, laboratory-specific intelligence, and reusable computational workflows.

The outcome

More reproducible scientific analysis and preserved institutional expertise.

Research DevelopmentExploring reusable scientific analytical methods.
EXPLORE A COLLABORATION

Expanding MATERIA’s laboratory collaborations.

MATERIA is being developed around real analytical challenges. We welcome conversations with laboratories, research institutions, and technology partners interested in exploring new analytical capabilities, developing laboratory-specific AI, and evaluating the platform within their own environments.

Laboratories · Research institutions · Microscopy manufacturers
Long-term vision

The intelligence layer for modern laboratories.

A connected software ecosystem in which laboratories develop, validate, preserve, improve and share analytical capabilities, so that scientific procedures become maintained computational infrastructure instead of isolated scripts and undocumented routines.

TODAYAlphaThe post-acquisition workflowIngest to export in one environment, with specialist inspection and traceable results.
NEXTIn developmentLaboratory-specific intelligenceWorkflow Builder, model development and controlled sharing of validated methods within an organization.
LONG TERMVisionAn ecosystem of analytical capabilitiesReusable methods, models and components that laboratories can publish and adopt, potentially through a marketplace.

Your laboratory generates data. MATERIA turns expertise into infrastructure.

Working on a repetitive microscopy analysis problem? Developing new analytical methods? Interested in technology integration? Let’s talk.

Explore the Platform