Multi-Modal Foundation · MahmoodLab
TITAN
Slide-level pathology foundation model generating whole-slide embeddings for oncology and biomarker prediction.
What TITAN is
TITAN is a computational pathology foundation model developed by MahmoodLab for slide-level analysis in medicine and biology. It generates holistic whole-slide embeddings to facilitate downstream diagnostic and predictive tasks from histology data.
The model is intended to assist researchers in studying tissue specimens, specifically addressing challenges in cancer subtyping, biomarker prediction, and treatment-response modelling.
How it works
TITAN is built on a Vision Transformer (ViT) architecture and trained using self-supervised learning techniques. It processes whole-slide histology images to extract high-dimensional visual feature representations at the slide level.
Key facts
- Architecture
- Vision Transformer (ViT)
- Learning paradigm
- Self-supervised learning
- Pipeline tag
- Image feature extraction
- Framework
- PyTorch
What you can use it for
- Cancer subtyping
- Deriving slide-level representations from tissue sections to classify and distinguish between histological cancer subtypes.
- Biomarker prediction
- Extracting whole-slide features to predict molecular and clinical biomarkers directly from pathology imagery.
- Treatment-response modelling
- Using slide embeddings to model and evaluate therapeutic responses in oncology workflows.
Run it here
We list this model but cannot run it yet: excluded (gated): Gated and CC-BY-NC-ND-4.0..
Everything on this page still applies. Running it needs a loader for this repository, not a bigger machine.
Strengths and limits
Good at
- Produces slide-level embeddings directly representing whole tissue sections.
- Employs self-supervised Vision Transformer representations transferable across multiple histology tasks.
Inputs and outputs
You provide
Image
PNG, JPEG or TIFF. Sent to the model as-is.
You get back
Mean-pooled representation, one vector per input.
Licence
Check the licence before commercial use · cc-by-nc-nd-4.0
This model is gated by its authors — its licence has to be accepted on the source repository before it will run.
Questions
Can I use TITAN commercially?
Check the licence before commercial use. This model is gated by its authors — its licence has to be accepted on the source repository before it will run.
What kind of architecture does TITAN use?
TITAN is implemented as a Vision Transformer (ViT) trained using a self-supervised approach.
What primary downstream applications does the model support?
The model is designed for cancer subtyping, biomarker prediction, and treatment-response modelling.
What output does TITAN generate?
It operates as an image feature extractor that outputs whole-slide embeddings from histology imagery.
Written from TITAN's own documentation (model card). Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.