Loading…
Longitudinal multimodal patient foundation model for oncology, fusing clinical records, DNA, RNA, and H&E pathology into one patient-state embedding.
Pan-cancer clinico-genomic model for treatment response and survival prediction, transferring zero-shot to unseen hospitals and cancer types.
Vision-language model for neuroblastoma pathology that reads H&E slides with their reports to grade tumors, infer biomarkers and stratify risk.
Transcriptome foundation model for precision oncology, generalizing zero-shot across tissue, plasma cfRNA, and tumor-educated platelet modalities.
BERT-style language model for somatic mutations, pretrained on cancer sequencing from 210,000+ patients for tumor subtyping and therapy response.
Multimodal foundation model that embeds histology, transcriptomics, and clinical records in one space for patient stratification and target discovery.
Cancer genomics foundation model embedding clinical gene-panel mutations into tumor subtype vectors. Pretrained on 30,328 tumors and 8 networks.
Whole-slide histopathology foundation model trained end-to-end on slide-level labels across 18 tasks, on 5% of the energy of SSL-trained peers.
PET/CT foundation model pretrained by cross-modal masked autoencoding on whole-body scans, for tumor lesion segmentation and lymphoma staging.
Semantic layout-guided 3D diffusion model that synthesizes thoracic CT volumes from a lung and nodule mask to expand lung cancer screening data.
Vision-language foundation model for precision oncology, pretrained on 50M pathology images and 1B text tokens via unified masked modeling.
Weakly supervised histopathology foundation model pretrained on 60,530 whole-slide images for cancer detection, prognosis, and molecular prediction.
Separable 4D CNN that strips rotational streak artifacts from respiration-resolved cone-beam CT, processing all ten breathing phases in one pass.
Self-supervised 3D CT foundation model that extracts general-purpose tumor representations for cancer imaging biomarker discovery and prognosis.
Sequence-only cancer driver mutation predictor combining ProtT5-XL embeddings with per-position evolutionary statistics and a confidence score.