Loading
TITAN is a model that summarises a whole slide, not just tiles, and can write a draft pathology report.
TITAN is a whole-slide pathology foundation model from the Mahmood Lab. Rather than embedding individual tiles, it runs a slide-level transformer over CONCH tile features and aligns the resulting representation with pathology report text, and it was pretrained on 335,645 whole slides using both vision-only and vision-language objectives. The slide-level embeddings support retrieval of rare cancers, prognosis prediction and generation of draft pathology reports, and it is released for research. Generated text needs pathologist review, and the model has not been prospectively validated in clinical workflows. Its contribution is architectural: it shows how to summarise a gigapixel slide into one vector that carries diagnostic meaning. TITAN is the step from reading patches of tissue to reading the whole slide.
Slide-level transformer over CONCH tile features, aligned with report text.
Query for this technology: (TITLE:"TITAN" OR ABSTRACT:"TITAN" OR TITLE:"whole-slide multimodal model" OR ABSTRACT:"whole-slide multimodal model") AND (cancer OR tumor OR tumour OR oncology OR carcinoma OR lymphoma OR leukemia OR leukaemia OR myeloma OR sarcoma OR melanoma OR glioma). Results are unfiltered search hits about TITAN (whole-slide multimodal model), not a curated reading list.
Shares Pathology & radiology foundation models, AI in oncology roadmap: pattern readers → foundation models → agents in the workflow, Dana-Farber Brigham Cancer Center and the tags foundation-model, pathology.
Shares AI in the oncology clinic: from narrow cleared tools to multimodal decision support, Pathology & radiology foundation models, AI in oncology roadmap: pattern readers → foundation models → agents in the workflow and the tags foundation-model, pathology.
Shares AI in the oncology clinic: from narrow cleared tools to multimodal decision support, Pathology & radiology foundation models, AI in oncology roadmap: pattern readers → foundation models → agents in the workflow and the tags foundation-model, pathology.
Shares AI in the oncology clinic: from narrow cleared tools to multimodal decision support, Pathology & radiology foundation models, AI in oncology roadmap: pattern readers → foundation models → agents in the workflow and the tags foundation-model, pathology.
Shares AI in the oncology clinic: from narrow cleared tools to multimodal decision support, Pathology & radiology foundation models and the tags foundation-model, pathology.
Shares Pathology & radiology foundation models, AI in oncology roadmap: pattern readers → foundation models → agents in the workflow and the tags foundation-model, pathology.
Shares Pathology & radiology foundation models and the tags foundation-model, pathology.
Shares Pathology & radiology foundation models and the tags foundation-model, pathology.
Open-source projects that implement or serve this technology, from OnCo's own catalogue: licence and last activity as the repository reported them on the day of the fetch. Listing is not endorsement; check the licence before reuse and the validation before clinical use.
A whole-slide foundation model that produces slide-level embeddings and reports, from the Mahmood lab.