Large-Scale Analysis

Analyze brain MRI at scale.

Use one approved research workspace to process MRI cases in the background and review structured report drafts.

For approved research projects. ReMIND is a research prototype, and every generated report requires qualified review.

Your data stays your data. Uploaded studies are used only to process your requested job. They are not used to train, fine-tune, or improve ReMIND.
COHORT_07 · 128 CASES Processing

Longitudinal MRI cohort

87 / 128 complete
CASE_0087 4 sequences Ready to review
CASE_0088 3 sequences Generating
CASE_0089 5 sequences Queued
CASE_0090 4 sequences Queued
NIfTI and DICOM studies
Multi-sequence cases
Resumable background jobs
Case-level reports and chat
One organized workspace

From dataset to case-level review.

Large-Scale Analysis brings the essential parts of a research batch workflow together without requiring teams to manage the model-serving infrastructure themselves.

01

Organize cohorts efficiently

Add cases manually or use a structured folder or ZIP archive with multiple MRI sequences per case.

02

Process in the background

Submit the job once, monitor case-level progress, and return later without keeping the browser open.

03

Review every result

Open each completed case to inspect its MRI sequences, report draft, and case-specific follow-up chat.

How it works

A clear path from access to analysis.

01 — APPLY

Describe your project

Submit the research purpose, institutional context, and expected dataset size.

02 — PREPARE

Organize by case

Use one folder per case, include one or more MRI sequences, and add optional context.

03 — PROCESS

Launch the batch job

ReMIND processes approved jobs in the background and records progress by case.

04 — REVIEW

Inspect the outputs

Review images and drafts, then ask follow-up questions within each case workspace.

Dataset preparation

One folder per case.

The immediate subfolder name becomes the Case ID. Each case can contain multiple NIfTI volumes or DICOM series.

Clinical context is optional

Include a TXT file in the case folder, enter context manually, or provide a CSV/XLSX mapping.

Keep identifiers out

Upload appropriately de-identified research data and do not submit patient information through the public application form.

Multiple sequences are supported

T1, T2, FLAIR, DWI, SWI, post-contrast, and other available sequences may be grouped within the same case.

dataset/
  CASE_001/
    T1.nii.gz
    T2.nii.gz
    FLAIR.nii.gz
    clinical_context.txt

  CASE_002/
    T1/
      DICOM files
    FLAIR/
      DICOM files

  clinical_context.csv
Responsible research use

Access and data controls by design.

Large-Scale Analysis is intended for governed institutional research—not consumer uploads or unsupervised clinical diagnosis.

Approved access

Batch Processing is limited to reviewed projects and authenticated users. Approval may depend on project fit, capacity, and data-governance requirements.

Project responsibility

Research teams remain responsible for institutional approval, permitted data use, appropriate de-identification, and removal of identifiers embedded in image pixels.

No model training

Uploaded studies and clinical context are not used to train, fine-tune, or improve ReMIND.

Protected job storage

Jobs are associated with their authenticated owner, stored in protected application storage, and expire according to the configured retention period.

Research boundary: Generated reports are drafts. Performance may vary across scanners, protocols, institutions, and populations, and the system is not a replacement for clinical interpretation.
FAQ

Before you apply.

What file formats are supported?

The workflow supports NIfTI files, DICOM files, ZIP archives, accepted JSON sidecars, and CSV/XLSX context mappings.

Will my data be used to train or improve ReMIND?

No. Uploaded studies and clinical context are used only to provide the requested batch-processing service and are not used to train, fine-tune, or improve the model.

Is clinical context required?

No. Context is optional and can be entered manually, stored in a TXT file, or provided through a CSV/XLSX mapping.

Can I leave and return to a running job?

Yes. Approved jobs run in the background, and authenticated users can return to active jobs after closing the browser.

How long are files and results retained?

Uploaded files, prepared files, and report drafts remain available for the configured job-retention period.

Is ReMIND intended for clinical diagnosis?

No. ReMIND is a research prototype, and all outputs require qualified review.

Start a project

See whether ReMIND fits your MRI study.

Tell us about your research project, dataset, and expected use. We will review the request and follow up through your institutional email. Your uploaded data will not be used for model training or improvement.