Request Access to Polaris™ Breast ROR Small v1
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
Polaris™ Breast ROR Small v1 is licensed under CC-BY-NC-ND 4.0 for non-commercial, academic research use only. Additional terms apply — please read the full Terms of Use on this page before submitting. Access is manually reviewed by StratifAI GmbH. By submitting this form you confirm you have read and agree to all terms set out on this page. Request access with your Current and official institutional email (this must match your primary email in your Hugging Face account,other domains will be denied). For research partnerships and collaborations contact info@stratifai.com.
Log in or Sign Up to review the conditions and access this model content.
Polaris™ Breast ROR Small v1.0
<<<<<<< HEAD Survival-Weibull model for invasive disease-free survival (iDFS) prediction with clinical covariates. To use Polaris™ Breast ROR v1.0 you must first have access to Polaris™ FMv2. The two models work together, Polaris™ FMv2 processes your slides first, and its outputs are then used by Polaris™ Breast ROR v1.0. Open source attribution for this model is provided in Polaris™ FMv2.
Survival-Weibull model for invasive disease-free survival (iDFS) prediction with clinical covariates. To use Polaris™ Breast ROR Small v1.0 you must first have access to Polaris™FMv2-small. The two models work together, Polaris™FMv2-small processes your slides first, and its outputs are then used by Polaris™ Breast ROR Small v1.0. Open source attribution for this model is provided in Polaris™FMv2-small.
This is the small variant: same architecture and training split as Polaris™ Breast ROR v1.0, but built on top of the 768-dim Polaris™FMv2-small features (with cat-max pooling) instead of the 1280-dim Polaris™FMv2-huge features. Use this variant when feature-extraction throughput or storage matters; use the huge variant for best validation C-index.
bcbece7 (Replace with small model)
LICENSE & TERMS OF USE
Please read the Terms of Use carefully before requesting access. By submitting your request you agree to be bound by them.
For clarity, Licensee = You and your institution.
License. This Model and associated code are released under the CC-BY-NC-ND 4.0 license and may only be used for non-commercial, academic research purposes with proper attribution. Any commercial use, sale, or other monetization of the Model and its derivatives, which include models trained on outputs from the Model or datasets created from the Model, is prohibited and requires prior approval. By downloading this model, you agree not to distribute, publish or reproduce a copy of the model. If another user within your organization wishes to use the Model, they must register as an individual user and agree to comply with the terms of use. Users may not attempt to re-identify the deidentified data used to develop the underlying Model.
No Ownership Transfer. This license does not constitute a transfer of ownership. StratifAI GmbH retains full ownership of the Model, its weights, architecture, all associated Intellectual Property, and related Know-How. The underlying DINOv2 and VIRCHOW components remain subject to their terms and the Apache 2.0 license and are the property of Meta Platforms, Inc.
Findings. As part of our research collaboration, we request that you share a summary of your findings with StratifAI before any public disclosure or publication. This helps us improve the Model and support your work.
Primary Use Cases
- Validation studies with vetted research and clinical partners
- Benchmarking and performance evaluation in computational pathology
Out-of-Scope Uses
- Clinical deployment without regulatory approval
- Any commercial use without a separate written agreement
DISCLAIMERS & WARRANTIES
The Model is provided "AS IS" without warranties of any kind, express or implied, including but not limited to warranties of merchantability, fitness for a particular purpose, or non-infringement. The Licensee acknowledges that the Model has not yet been approved by regulatory authorities for any clinical use, and that it is provided without any warranty or guaranteed availability. The Model is available for Research Use Only (RUO).
Model outputs do not constitute professional or medical advice, including as a substitute for clinical judgment, diagnosis, treatment, or prevention of disease.
REVOCATION OF ACCESS
Revocation for Breach. StratifAI GmbH reserves the right to immediately revoke access to the Model, without prior notice, in the event of any breach of these terms
Contact
| Purpose | Contact |
|---|---|
| Access & research partnerships | info@stratifai.com |
| Legal & licensing enquiries | legal@stratifai.com |
| Press & marketing | marketing@stratifai.com |
Model details
- Target: iDFS_time (survival-weibull) <<<<<<< HEAD
Feature extractor: Polaris™ FMv2-huge (
StratifAI/PolarisFMv2-huge, 1280-dim)- Feature extractor: Polaris™FMv2-small with
cat-maxpooling (StratifAI/PolarisFMv2-small, 768-dim)bcbece7 (Replace with small model)
- Architecture: multimodal_transformer (WSI + 8 clinical covariates)
Clinical covariates
| Covariate | Type | Values |
|---|---|---|
| ER_status | categorical | 0, 1 |
| PR_status | categorical | 0, 1 |
| HER2_status | categorical | 0, 1 |
| Menopausal_status | categorical | Pre, Peri, Post |
| T_stage | categorical | 0, 1, 2, 3, 4 |
| N_stage | categorical | 0, 1, 2, 3 |
| Histo_grade | categorical | 1, 2, 3 |
| Age_surgery | continuous | mean=53.98, std=11.74 |
Sample clinical CSV
Px_UID,ER_status,PR_status,HER2_status,Menopausal_status,T_stage,N_stage,Histo_grade,Age_surgery
PAT-001,1,1,0,Post,2,1,2,55
PAT-002,1,1,1,Post,1,0,3,50
PAT-003,1,0,0,Pre,2,1,,62
All columns are optional - missing or empty values are handled gracefully. Refer to the covariate table above for accepted values.
Usage
This model is available on HuggingFace:
sai-predict \
--model-url https://huggingface.co/StratifAI/polaris-breast-ror-small \
--feature-dir /path/to/polarisfm-small-cat-max-features \
--clini /path/to/clini.csv \
--slide-csv /path/to/slide.csv \
--output ./predictions.json
Or using local config/weights:
sai-predict \
--model-config ./config.json \
--weights ./weights.pt \
--feature-dir /path/to/polarisfm-small-cat-max-features \
--clini /path/to/clini.csv \
--slide-csv /path/to/slide.csv \
--output ./predictions.json
Features must be extracted with StratifAI/PolarisFMv2-small using cat-max pooling (768-dim, tile_size_px=224, tile_size_um=256). Clinical covariates (ER, PR, HER2, menopausal status, T/N stage, histological grade, age at surgery) may be provided via --clini; missing values are handled by masked attention.
How this model was created
Source
<<<<<<< HEAD Trained via the MCP pipeline using Polaris™ FMv2-huge (224px/224µm) features from 3 cohorts, 5-fold cross-validation.
Trained via the MCP train_survival_url_ pipeline (sai-engine-inference v0.5.3), using Polaris™FMv2-small (224px/224µm, cat-max pooling) features from 3 cohorts. Same fold split as Polaris™ Breast ROR v1.0.
bcbece7 (Replace with small model)
Data
| Split | N | Events |
|---|---|---|
| Train | 3813 | — |
| Valid | 956 | — |
Model architecture
multimodal_transformer v1.0 — combining WSI and clinical pathways:
- Outer model: dim=99, heads=11, layers=3, dropout=0.02, wsi_dropout=0.25, clinical_data_dropout=0.25
- Base WSI model (
masked_vision_transformerv1.0): dim_input=768, dim_model=180, mlp_dim=360, heads=3, layers=2, dropout=0.11
Checkpoint selection
Top-3 by cindex_valid_window:
| Step | valid_cindex |
|---|---|
| 1200 | 0.6676 ← packaged |
| 600 | 0.6661 |
| 0 | 0.5000 (random baseline) |
Training early-stopped at step 9000; validation peaked at step 1200, after which train/valid diverged (overfit). The packaged weights are the step-1200 top-k checkpoint.
Model card last updated: 04.05.2026 — Polaris™ Breast ROR Small v1 StratifAI GmbH Karl-Liebknecht-Str. 29A, 10178 Berlin, Germany stratifai.com
- Downloads last month
- 15