Frontier vision-language AI for neurology

The clinical reasoning layer for brain MRI.

ReMIND reads multi-sequence brain MRI the way a neuroradiologist would — reasoning across T1, T2, FLAIR, DWI and beyond to draft structured, radiology-style reports.

RESEARCH PROTOTYPE — NOT FOR CLINICAL USE. Outputs are drafts intended to support review by qualified clinicians.
STUDY_001 · multi-sequence
Reasoning
T1
T2
FLAIR
DWI
Findings · Draft
Confluent T2/FLAIR hyperintensities in the periventricular white matter, consistent with chronic small-vessel ischemic changes. Mild generalized volume loss is age-appropriate.
Modality-faithful · 0 unsupported claims
Trained & validated across
Gradient Health
Boston Medical Center
Lahey Hospital & Medical Center
Brain & Spine Hospital · Chennai
National Alzheimer's Coordinating Center
70K+
multi-sequence
MRI visits
850K+
paired imaging
sequences
1M+
grounded clinical
question–answer pairs
5
cohorts across
multiple countries
Usage signal

How ReMIND is being used.

Approximate geographic reach from signed-in app visits. Locations are aggregated and used only to show adoption of our research tool.

Studies analyzed
avg time / report
tracked sessions
locations
Waiting for city-level session data.
How it works

Upload a study.
Get a draft. Ask anything.

ReMIND was built around how neuroradiologists actually read — synthesizing complementary contrasts into a coherent narrative, then producing structured language a clinician or a researcher can edit and trust.

01 — UPLOAD

Bring a multi-sequence study

DICOM and NIfTI in. T1, T2, FLAIR, DWI, SWI, post-contrast, MRA/MRV — any combination, any protocol.

02 — REASON

Cross-sequence interpretation

A vision-language model jointly attends across sequences, surfacing focal and diffuse findings the way an integrated read would.

03 — REVIEW

Draft report & conversation

A structured radiology-style draft you can edit, plus a chat to interrogate findings, sequences and differentials.

What's inside

Specialized depth, not generic scale.

Frontier-grade behavior on neuroimaging without frontier-scale compute. Every component is tuned to how brain MRI is acquired, read and reported.

Multi-sequence reasoning

Joint understanding across T1, T2, FLAIR, DWI/ADC, SWI/T2*, post-contrast, and angiographic series — without explicit cueing.

cross-contrast

Radiology-style structured reports

Drafts mapped to the canonical seven sections — Examination, Technique, Comparison, History, Indication, Findings, Impression.

clinician-ready

Conversational case Q&A

Ask follow-ups across various question types — from yes/no verification to free-form clinical reasoning grounded in the scan.

grounded VQA
Validation

Tested where the data actually lives.

ReMIND was developed on a large collection of clinical cases and held out on independent ones — across academic, community, and international settings — so generalization isn't measured on the data it was trained on.

Train

Gradient Health

Multi-site clinical · USA
65,538
visits
External

Boston Medical Center

Boston, MA · USA
3,314
visits
External

Lahey Hospital & Medical Center

Burlington, MA · USA
2,100
visits
External

Brain & Spine Hospital

Chennai · India
388
visits
Research

National Alzheimer's Coordinating Center

Multi-site · USA
2,286
visits
A point of view

In high-stakes medical imaging, specialization outperforms general-purpose scale. Domain-specific tuning and rigorous external validation are how AI earns the right to assist a clinician — and that's the bar ReMIND was built to meet.

— Kolachalama Laboratory
Our commitments

Open science, responsibly.

Open methodology

Architecture, data pipeline, decoding strategy and metrics are published. We welcome scrutiny — that's what peer review is for.

Clinician oversight

Every output is labeled a draft. The model proposes; the clinician decides. ReMIND assists, it doesn't replace.

Equity by design

Validated across academic, community and international cohorts so performance isn't only a story about one population.

Privacy-first handling

Studies are de-identified and handled with strict safeguards. No data is stored on our servers after the user session ends.

Try it now

Bring a brain MRI. See what ReMIND reads.

The demo runs on de-identified studies. No friction — just a draft you can interrogate, and a clear boundary about what it is and isn't.