Leakage-free foundation-model (UNI2-h + Virchow2 + Phikon-v2) + chunked-XGBoost ensemble for 10-class brain-tumor histopathology patch classification — BraTS 2026 PATH. OOF acc 0.762 / MCC 0.713.
-
Updated
Jun 8, 2026 - Python
Leakage-free foundation-model (UNI2-h + Virchow2 + Phikon-v2) + chunked-XGBoost ensemble for 10-class brain-tumor histopathology patch classification — BraTS 2026 PATH. OOF acc 0.762 / MCC 0.713.
Reliability analysis for MICCAI BraTS-GoAT 2026, comparing single-model confidence against deep-ensemble disagreement on calibration and error detection under graded synthetic acquisition shift.
NeuroTS-Net is a three-dimensional, multi-class semantic segmentation neural network architecture designed for pediatric brain tumor segmentation in multimodal MRI
Team HMNUnet's BraTS-PEDs 2026 Task 2 submission: two frozen nnU-Net ensembles plus a budgeted, invariant-preserving ET transplant
To associate your repository with the brats2026 topic, visit your repo's landing page and select "manage topics."