Systems
Modular biomedical software components for evidence review, workflow pilots, model evaluation, and technical due diligence.
TrialAI-Lab
Clinical-trial intelligence: completion risk, feasibility, eligibility criteria NLP, adverse-event signals, trial similarity, and time-to-completion analysis.
BioGraph-AI
Executable fixed-graph disease-target ranking with Open Targets evidence components, named drug bridges, and a prominently weak release holdout.
GliomaVision
Medical imaging workflows for brain-tumor MRI: classification, segmentation, grading, uncertainty, interpretability, and robustness under dataset or scanner variation.
MIMIC-RiskLab
Aggregate open-demo audit of cohort construction, leakage controls, calibration instability, subgroup sample sizes, reproduction drift, and credentialed-data blockers—without patient-level prediction.
SafeRL-ICU
Aggregate off-policy-evaluation audit of estimator disagreement, importance-weight diagnostics, sensitivity, and blockers—without policy execution or treatment advice.
MedAI Eval
Shared evaluation layer across MedCoria systems: classification metrics, calibration analysis, reliability curves, subgroup evaluation, robustness checks, and model-card generation.