Search papers, labs, and topics across Lattice.
This study investigates the use of CT-CLIP representations as a feature extractor for multimodal survival prediction in lung cancer, addressing the challenge of limited curated imaging data. By evaluating various adaptation strategies, including frozen encoders and low-rank adaptation, the researchers demonstrate that a frozen CT-CLIP model paired with a lightweight survival head surpasses traditional clinical baselines. The findings indicate that this approach effectively stratifies patients into high- and low-risk categories, enhancing prognostic accuracy in data-constrained environments.
A frozen CT-CLIP model can outperform traditional clinical baselines in lung cancer survival prediction, even with limited data.
Accurate prognosis prediction is important for treatment planning in lung cancer, but deep learning-driven survival modelling is often limited by the scarcity of curated imaging cohorts with reliable outcome data. This study evaluates whether representations from a domain-specific foundation model can be used for multimodal survival prediction in data-constrained clinical settings. We assess the foundation model CT-CLIP as a feature extractor for pretreatment computed tomography images and clinical variables from 242 diagnosed lung cancer patients. The evaluation includes adaptation strategies based on frozen encoders, full fine-tuning, and low-rank adaptation, together with modality ablations and comparisons with clinical and multimodal baselines. The results show that a frozen CT-CLIP model combined with a trainable lightweight survival head outperforms the clinical baseline and achieves comparable or improved performance relative to other multimodal approaches, and separates patients into clinically meaningful high- and low-risk groups.
