Time-Conditioned and Multi-Time Survival Prediction from 2D PET/CT Projections in Lung Cancer

2026-06-10

AI Summary

This study addresses the challenge of accurately predicting overall survival in non-small cell lung cancer patients from PET/CT imaging and systematically evaluates the impact of temporal modeling on predictive performance. To this end, two novel approaches are proposed: the Attention-guided Temporal Conditional Survival model (ATCS) and the Multi-temporal Survival model (MTS). ATCS leverages a temporal conditioning mechanism, while MTS explicitly models multiple time points; both integrate tumor-specific and tissue-level PET/CT features to differentially capture short- and long-term survival risks. Evaluated via five-fold cross-validation on a cohort of 848 patients, ATCS and MTS achieve mean time-dependent AUCs of 0.794 and 0.793, respectively, significantly outperforming the baseline Temporal Conditional Survival (TCS) model (AUC = 0.767), with each demonstrating complementary strengths across distinct follow-up intervals.

Abstract

Accurate prediction of overall survival (OS) from positron emission tomography/computed tomography (PET/CT) can support personalized treatment and follow-up strategies in oncology. However, the impact of temporal modeling on imaging-based survival prediction remains insufficiently explored. We investigate how different temporal formulations influence survival prediction by developing two complementary approaches: Attention-guided Time-Conditioned Survival (ATCS) and Multi-Time Survival (MTS). We retrospectively analyzed pre-treatment PET/CT images from 848 patients with non-small cell lung cancer (NSCLC), including 556 for model development and 292 for held-out testing. A previously proposed Time-Conditioned Survival (TCS) model was used as a baseline. Models were trained using 5-fold cross-validation and evaluated on the test set using time-dependent area under the curve (AUC) at 6-month intervals from 0.5 to 5 years. Both ATCS and MTS outperformed the baseline TCS model, achieving mean AUCs of 0.794 and 0.793, respectively, compared to 0.767. ATCS performed better at earlier time points (0.5-3 years), whereas MTS performed better at later intervals (3.5-5 years). Combining tumor-specific and tissue-wise PET/CT features improved performance over either input alone. Finer temporal discretization improved short-term prediction, while coarser intervals provided more stable long-term estimates. These findings demonstrate that temporal modeling and input design influence PET/CT-based survival prediction. The proposed approaches enable time-specific survival estimation from pre-treatment imaging and may support improved risk stratification and clinical decision-making.

Problem

• survival prediction
• PET/CT
• temporal modeling
• lung cancer
• time-conditioned

Innovation

• Time-Conditioned Survival
• Multi-Time Survival
• Attention Mechanism
• Temporal Modeling
• PET/CT-based Prediction