QuantBrain · AI-Native Brain Quantification Platform
Automatic ventricle segmentation and volumetrics for hydrocephalus and shunt management
Handles both CT and MRI (T1, T1c, T2, FLAIR) — the scans NPH workups, shunt follow-ups, and pediatric exams routinely use, where most tools are MRI-only. Computes Evans Index from a 3-D mask, not a hand-picked slice, and flags every output with a per-scan uncertainty score.
What it does
Segment, measure, follow
Segment the ventricles
- Fully-automatic lateral-ventricle segmentation across CT and T1, T2, and FLAIR MRI.
- Editable contours when the auto-segmentation needs adjustment.
Measure what matters
- Evans Index computed from the 3-D mask, not a hand-picked slice.
- Per-region volume in cm³ with a built-in uncertainty score on every output.
Follow patients over time
- Side-by-side comparison viewer, including across CT and MRI timepoints.
- Volume and Evans-Index trend — the question that drives shunt decisions.
Validation & fit
Numbers, indications, and where it runs
0.92
Mean Dice — CT
0.90
Mean Dice — MRI
Across T1, T2, and FLAIR.
4
Disease populations validated
Normal, hydrocephalic, tumor, shunted — pediatric and adult.
Intended use. QuantBrain — Ventricle is intended for radiologists, neurologists, neuro-radiologists, pediatricians, and other physicians as an initial machine-learning approach for ventricle segmentation to support diagnosis and management of hydrocephalus and shunt malfunction. Outputs are reviewed by a qualified clinician.
Modalities
CT · MRI (T1, T1c, T2, FLAIR) · pediatric and adult
Inputs & outputs
DICOM in (push, query, upload); NIfTI / NRRD also accepted; DICOM Seg, CSV, PDF out
Deployment
Web-based, on-premise behind your firewall · Batch CLI for cohort research
Publications
Peer-reviewed validation and clinical use
-
Morphological changes of the choroid plexus in the lateral ventricle across the lifespan: 5551 subjects from fetus to elderly
-
Prediction of shunt malfunction using automated ventricular volume analysis and radiomics
-
MRI-based prediction of clinical improvement after ventricular shunt placement for normal pressure hydrocephalus
-
Associations between the choroid plexus and tau in Alzheimer's disease using an active learning segmentation pipeline
-
Establishment of age- and sex-specific reference cerebral ventricle volumes
-
Automated 3D fetal brain segmentation using an optimized deep learning approach
Quantify ventricles on your own data
Research use today; 510(k) submission in preparation.