Further processing options
available via online resource

Long-term cancer survival prediction using multimodal deep learning

Saved in:

Published in: Scientific reports pages:1-12; 11(2021), Artikel-ID 13505, Seite 1-12; volume:11; year:2021; elocationid:13505; extent:12
Authors and Corporations: Vale Silva, Luis A. (Author), Rohr, Karl (Author)
Other Authors: Rohr, Karl [Author]
Type of Resource: E-Book Component Part
Language: English
29 June 2021
Series: Scientific reports, 11(2021), Artikel-ID 13505, Seite 1-12
Source: Verbunddaten SWB
Lizenzfreie Online-Ressourcen
ISSN: 2045-2322
Summary: The age of precision medicine demands powerful computational techniques to handle high-dimensional patient data. We present MultiSurv, a multimodal deep learning method for long-term pan-cancer survival prediction. MultiSurv uses dedicated submodels to establish feature representations of clinical, imaging, and different high-dimensional omics data modalities. A data fusion layer aggregates the multimodal representations, and a prediction submodel generates conditional survival probabilities for follow-up time intervals spanning several decades. MultiSurv is the first non-linear and non-proportional survival prediction method that leverages multimodal data. In addition, MultiSurv can handle missing data, including single values and complete data modalities. MultiSurv was applied to data from 33 different cancer types and yields accurate pan-cancer patient survival curves. A quantitative comparison with previous methods showed that Multisurv achieves the best results according to different time-dependent metrics. We also generated visualizations of the learned multimodal representation of MultiSurv, which revealed insights on cancer characteristics and heterogeneity.
Item Description: Gesehen am 12.08.2021
Physical Description: 12
ISSN: 2045-2322
DOI: 10.1038/s41598-021-92799-4