Organize cohorts efficiently
Add cases manually or use a structured folder or ZIP archive with multiple MRI sequences per case.
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.
Large-Scale Analysis brings the essential parts of a research batch workflow together without requiring teams to manage the model-serving infrastructure themselves.
Add cases manually or use a structured folder or ZIP archive with multiple MRI sequences per case.
Submit the job once, monitor case-level progress, and return later without keeping the browser open.
Open each completed case to inspect its MRI sequences, report draft, and case-specific follow-up chat.
Submit the research purpose, institutional context, and expected dataset size.
Use one folder per case, include one or more MRI sequences, and add optional context.
ReMIND processes approved jobs in the background and records progress by case.
Review images and drafts, then ask follow-up questions within each case workspace.
The immediate subfolder name becomes the Case ID. Each case can contain multiple NIfTI volumes or DICOM series.
Include a TXT file in the case folder, enter context manually, or provide a CSV/XLSX mapping.
Upload appropriately de-identified research data and do not submit patient information through the public application form.
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
Large-Scale Analysis is intended for governed institutional research—not consumer uploads or unsupervised clinical diagnosis.
Batch Processing is limited to reviewed projects and authenticated users. Approval may depend on project fit, capacity, and data-governance requirements.
Research teams remain responsible for institutional approval, permitted data use, appropriate de-identification, and removal of identifiers embedded in image pixels.
Uploaded studies and clinical context are not used to train, fine-tune, or improve ReMIND.
Jobs are associated with their authenticated owner, stored in protected application storage, and expire according to the configured retention period.
The workflow supports NIfTI files, DICOM files, ZIP archives, accepted JSON sidecars, and CSV/XLSX context mappings.
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.
No. Context is optional and can be entered manually, stored in a TXT file, or provided through a CSV/XLSX mapping.
Yes. Approved jobs run in the background, and authenticated users can return to active jobs after closing the browser.
Uploaded files, prepared files, and report drafts remain available for the configured job-retention period.
No. ReMIND is a research prototype, and all outputs require qualified review.
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.