Contribution of imaging-genetics to overall survival prediction compared to clinical status for PCNSL patients

Amine Rebei (1) , Agusti Alentorn (2, 3) , Hamza Chegraoui (1) , Vincent Frouin (1) , Cathy Philippe (1)

1 BAOBAB - Building large instruments for neuroimaging: from population imaging to ultra-high magnetic fields
2 ICM - Institut du Cerveau = Paris Brain Institute
3 CHU Pitié-Salpêtrière [AP-HP]

Accurately predicting the survival of patients with cancer has the potential to substantially enhance and customize the treatment strategies. Integrating and using all the patients' available data is essential to make the most accurate predictions. In this work, we gather clinical, imaging and genetic data into one mono-block multivariate survival analysis for patients with primary central nervous system lymphoma (PC-NSL). As a first step, we select the best features from each pre-processed dataset. Then we assemble and use the resulting block to predict overall survival with a survival random forest algorithm. The assessment of the proposed method yielded a C-index of 0.776. We thus conclude that multimodal data integration significantly improves prediction performance.

Mots clés: Survival analysis; imaging genetics; MRI; Brain Tumor

Domaines: Machine Learning [stat.ML]; Imagerie; Cancer; Génétique humaine

Amine Rebei, Agusti Alentorn, Hamza Chegraoui, Vincent Frouin, Cathy Philippe. Contribution of imaging-genetics to overall survival prediction compared to clinical status for PCNSL patients. IEEE ISBI 2021 - International Symposium on Biomedical Imaging, IEEE, Apr 2021, Nice, France. ⟨cea-03162611⟩