MOOZY: A Patient-First Foundation Model for Computational Pathology

MOOZY is a slide and patient-level foundation model for computational pathology. The patient case, not the individual slide, is the core unit of representation. A vision-only slide encoder pretrained with masked self-distillation on 77,134 public slides is aligned with clinical semantics through multi-task supervision over 333 tasks (205 classification, 128 survival) from 56 public datasets spanning 23 anatomical sites. A case transformer explicitly models dependencies across all slides from the same patient, replacing the naive early/late fusion used by prior methods. 85.77M total parameters. Trained entirely on public data.

Table of Contents

• Installation
• Usage
  • From pre-computed H5 feature files
  • From raw whole-slide images
  • Python API
  • Arguments
  • Output format
• Architecture
• Tasks
• Citation
• License

Installation

pip install moozy

The checkpoint and task definitions are downloaded automatically from this repository on first use.

Usage

From pre-computed H5 feature files

The faster path. Pass .h5 files containing patch features extracted with lunit_vit_small_patch8_dino at 224x224 patch size. Compatible with AtlasPatch and TRIDENT outputs.

moozy encode slide_1.h5 slide_2.h5 --output case_embedding.h5

From raw whole-slide images

Pass slide files directly (.svs, .tiff, .ndpi, .mrxs, etc.). MOOZY calls AtlasPatch under the hood to segment tissue, extract patches, and compute features. Requires atlas-patch, sam2, and the OpenSlide system library (see the AtlasPatch installation guide).

moozy encode slide_1.svs slide_2.svs --output case_embedding.h5 --target_mag 20

Python API

from moozy.encoding import run_encoding

# From H5 feature files
run_encoding(
    slide_paths=["slide_1.h5", "slide_2.h5"],
    output_path="case_embedding.h5",
)

# From raw slides
run_encoding(
    slide_paths=["slide_1.svs", "slide_2.svs"],
    output_path="case_embedding.h5",
    target_mag=20,
)

Arguments

Output format

The output H5 file contains a features dataset (768-D float32 case embedding) and a coords dataset with slide metadata.

Architecture

Tasks

This repository includes 497 task definitions in the tasks/ directory. Each task has a config.yaml (task type, organ, label mapping) and a task.csv (annotations and splits). MOOZY was trained on 333 of them (205 classification and 128 survival endpoints) across all 32 TCGA cohorts, all 10 CPTAC cohorts, REG, BC-Therapy, BRACS, CAMELYON17, DHMC Kidney, DHMC LUAD, EBRAINS, IMP Colorectum, IMP Cervix, MBC, MUT-HET-RCC, NADT Prostate, NAT-BRCA, and PANDA. The remaining 164 tasks are additional TCGA annotations (AJCC pathologic staging and molecular subtypes), and MOOZY was not trained on any of them.

Citation

@misc{kotp2026moozypatientfirstfoundationmodel,
      title={MOOZY: A Patient-First Foundation Model for Computational Pathology},
      author={Yousef Kotp and Vincent Quoc-Huy Trinh and Christopher Pal and Mahdi S. Hosseini},
      year={2026},
      eprint={2603.27048},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2603.27048},
}

License

CC BY-NC-SA 4.0. Research and non-commercial use only.