Predicting the likelihood of survival is of paramount im-
portance for individuals diagnosed with cancer as it provides invalu-
able information regarding prognosis at an early stage. This knowledge
enables the formulation of effective treatment plans that lead to im-
proved patient outcomes. In the past few years, deep learning mod-
els have provided a feasible solution for assessing medical images, elec-
tronic health records, and genomic data to estimate cancer risk scores.
However, these models often fall short of their potential because they
struggle to learn regression-aware feature representations. In this study,
we propose Survival Rank-N-Contrast (SurvRNC) method, which intro-
duces a loss function as a regularizer to obtain an ordered represen-
tation based on the survival times. This function can handle censored
data and can be incorporated into any survival model to ensure that
the learned representation is ordinal. The model was extensively eval-
uated on a HEad & NeCK TumOR (HECKTOR) segmentation and
the outcome-prediction task dataset. We demonstrate that using the
SurvRNC method for training can achieve higher performance on dif-
ferent deep survival models. Additionally, it outperforms state-of-the-art
methods by 3.6% on the concordance index.
Paper Link: https://arxiv.org/pdf/2403.10603