Safetensors

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this model content.

Histopathology Slide Vector Quantized Variational Auto-Encoder (HSVQVAE)

We present the Histopathology Slide Vector Quantized Variational Autoencoder (HSVQVAE), a deep generative model designed for representation learning on whole‑slide histopathology images. HSVQVAE leverages the Vector Quantized Variational Autoencoder (VQ‑VAE) framework to capture rich, discrete latent representations of tissue morphology. By encoding slides into a compressed latent space, the model facilitates downstream tasks such as classification, clustering, and retrieval of histopathological patterns.

The model was trained on The Cancer Genome Atlas (TCGA) dataset, incorporating multiple views of histopathology slides to enhance robustness and generalizability. This multi‑view training strategy enables HSVQVAE to learn consistent representations across different magnifications and tissue contexts, reflecting the way pathologists examine slides at varying scales.

HSVQVAE provides a foundation for research in computational pathology, offering a reusable and interpretable representation space that can accelerate the development of diagnostic and prognostic tools. The model is released for research purposes only, with the aim of supporting academic exploration into representation learning for digital pathology.

License and Intended Use

This model is released under the CC-BY-NC 4.0 License (Attribution–NonCommercial). It is intended for research and academic purposes only. Commercial use of this model or its derivatives is strictly prohibited.

Disclaimer

This model is provided "as is" without any warranties or guarantees of any kind. The authors and contributors assume no responsibility or liability for any use, misuse, or potential harms arising from the application of this model. Users are solely responsible for ensuring that their use of the model complies with applicable laws, ethical standards, and institutional policies.

Citation

If you use this model in your research, please cite it as:

Yu Ando, Seokhwan Ko, Ji Young Park, Junghwan Cho, and Hyungsoo Han. HSVQVAE. Hugging Face, 2025. DOI: 10.57967/hf/6603.

Updates

  • [2025-09-27] Added model parameters and model files.
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support