Radiomics improves kidney cancer classification by adjusting renaclIP image features
Radiomics-Conditioned Modulation of RenalCLIP Features for Clear Cell Renal Cell Carcinoma Classification
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Classifying kidney tumors accurately from CT scans is difficult with small labeled datasets. The authors combined traditional radiomics, which extracts detailed tumor information from images, with RenalCLIP features, derived from a specialized model, to improve classification. They used a technique called feature-wise linear modulation (FiLM) to let radiomics adjust the RenalCLIP features, plus they added radiomics information directly. This combined approach worked better than other methods at identifying kidney cancer in two test groups.
What this means in practice
- •For medical imaging software developers: Create CT scan analysis tools that combine radiomics and RenalCLIP features using FiLM to improve kidney tumor classification accuracy.
- •For oncology data teams: Improve diagnostic pipelines for kidney cancer by integrating radiomics-modulated RenalCLIP features, enhancing classification reliability on small datasets.
Authors
Yuan Liang, Sourav Bhattacharjee, Abraham Campbell
Abstract
Radiomics provides quantitative descriptions of tumour appearance that may complement disease-specific foundation models in small labelled cohorts. We investigate this complementarity for computed tomography-based classification of clear cell renal cell carcinoma. Our framework uses radiomics to modulate RenalCLIP features through feature-wise linear modulation (FiLM), while retaining a direct radiomics contribution. Internal testing and external validation compare it with conventional fusion strategies and reference classifiers. The FiLM model achieves an area under the receiver operating characteristic curve (AUC) of 0.804 internally and 0.854 externally, with the highest mean AUC among the evaluated RenalCLIP fusion strategies in both cohorts. Pathway ablations examine the contributions of conditional modulation and the direct radiomics residual, while feature permutation highlights the role of tumour texture. These findings support radiomics as a useful complement to RenalCLIP in a small labelled cohort and identify FiLM as an effective approach to integrating their representations for robust renal tumour classification.