SwarmDeepSurv: swarm intelligence advances deep survival network for prognostic radiomics signatures in four solid cancers.

Survival models exist to study relationships between biomarkers and treatment effects. Deep learning-powered survival models supersede the classical Cox proportional hazards (CoxPH) model, but substantial performance drops were observed on high-dimensional features because of irrelevant/redundant information. To fill this gap, we proposed SwarmDeepSurv by integrating swarm intelligence algorithms with the deep survival model. Furthermore, four objective functions were designed to optimize prognostic prediction while regularizing selected feature numbers. When testing on multicenter sets (n = 1,058) of four different cancer types, SwarmDeepSurv was less prone to overfitting and achieved optimal patient risk stratification compared with popular survival modeling algorithms. Strikingly, SwarmDeepSurv selected different features compared with classical feature selection algorithms, including the least absolute shrinkage and selection operator (LASSO), with nearly no feature overlapping across these models. Taken together, SwarmDeepSurv offers an alternative approach to model relationships between radiomics features and survival endpoints, which can further extend to study other input data types including genomics.

Informations de copyright

© 2023 The Author(s).

Déclaration de conflit d'intérêts

B.A.-L. declares commercial interest in Exscientia and AstraZeneca and is/was a consultant/scientific advisory board member for GlaxoSmithKline (GSK), Open Targets, Astex Pharmaceuticals, and Astellas Pharma, and is an ex-employee of Inpharmatica Ltd., all outside of the submitted work. J.Y.C. has received travel sponsorship from Accuray and Varian MedicalSystems and grants from Varian Medical Systems, outside the submitted work. N.I.V. receives consulting fees from Sanofi, Regeneron, Oncocyte, and Eli Lilly and research funding from Mirati, outside the submitted work. J.Z. reports research funding from Merck, Johnson & Johnson, and Novartis and consultant fees from Bristol Myers Squibb (BMS), Johnson & Johnson, AstraZeneca, Geneplus, OrigMed, Novartis, and Innovent, outside the submitted work. J.V.H. serves on scientific advisory boards for AstraZeneca, Boehringer Ingelheim, Genentech, GSK, Eli Lilly, Novartis, Spectrum, EMD Serono, Sanofi, Takeda, Mirati Therapeutics, BMS, and Janssen Global Services; receives research support from AstraZeneca, Takeda, Boehringer Ingelheaim, and Spectrum; and receives licensing fees from Spectrum, all outside of the submitted work. The other authors declare no competing interests in the submitted work.