SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.00224 · ICML · 2024

Multimodal Prototyping for cancer survival prediction

Guillaume Jaume, Richard Chen, Faisal Mahmood, Andrew Song, Anurag Vaidya, Alexander Baras

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 12 functions out of this paper's own repositories and ran 7 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
mahmoodlab/MMP — 7 of 12
FunctionStatusWhere it lives
CoxLoss Ran mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("05faa7f292aabeb5")
FeedForward Ran mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("a206b906c4df0eba")
NLLSurvLoss Ran mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("71951ad3b92ea5b9")
SNN_Block Ran mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("10482673211edd0c")
SurvRankingLoss Ran mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("95489d9d0796dea4")
init_per_path_model Ran mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("44e13a864df9f026")
partial_ll_loss Ran mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("4dd75c9172f12cfb")
MMAttention Not yet run mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("721d2d0c102ab4ae")
MMAttentionLayer Not yet run mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("323ddf11c2d03fca")
SurvPath Not yet run mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("b3e9ddc34af4aad1")
nll_loss Not yet run mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("15821d5a95c321e7")
process_surv Not yet run mahmoodlab/MMP/src/mil_models/model_multimodal.py
pointer only (licence: NOASSERTION) · get_code("64e4caddca1b2fb3")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Multimodal survival methods combining gigapixel histology whole-slide images (WSIs) and transcriptomic profiles are particularly promising for patient prognostication and stratification. Current approaches involve tokenizing the WSIs into smaller patches (> 10 4 patches) and transcriptomics into gene groups, which are then integrated using a Transformer for predicting outcomes. However, this process generates many tokens, which leads to high memory requirements for computing attention and complicates post-hoc interpretability analyses. Instead, we hypothesize that we can: (1) effectively summarize the morphological content of a WSI by condensing its constituting tokens using morphological prototypes, achieving more than 300× compression; and (2) accurately characterize cellular functions by encoding the transcriptomic profile with biological pathway prototypes, all in an unsupervised fashion. The resulting multimodal tokens are then processed by a fusion network, either with a Transformer or an optimal transport cross-alignment, which now operates with a small and fixed number of tokens without approximations. Extensive evaluation on six cancer types shows that our framework outperforms state-of-the-art methods with much less computation while unlocking new interpretability analyses. The code is available at https: //github.com/mahmoodlab/MMP.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2407.00224")
get_code_for_paper("2407.00224")
have("2407.00224")

Connect an agent — have() is free.