Dynamic Voting in Multi-view Learning for Radiomics Applications

Hongliu Cao (1, 2) , Simon Bernard (1) , Laurent Heutte (1) , Robert Sabourin (2)

1 DocApp - LITIS - Equipe Apprentissage
2 ETS - Ecole de Technologie Supérieure [Montréal]

Cancer diagnosis and treatment often require a personalized analysis for each patient nowadays, due to the heterogeneity among the different types of tumor and among patients. Radiomics is a recent medical imaging field that has shown during the past few years to be promising for achieving this personalization. However, a recent study shows that most of the state-of-the-art works in Radiomics fail to identify this problem as a multi-view learning task and that multi-view learning techniques are generally more efficient. In this work, we propose to further investigate the potential of one family of multi-view learning methods based on Multiple Classifier Systems where one classifier is learnt on each view and all classifiers are combined afterwards. In particular, we propose a random forest based dynamic weighted voting scheme, which personalizes the combination of views for each new patient to classify. The proposed method is validated on several real-world Radiomics problems.

Mots clés: Radiomics; dissimilarity; random forest; dynamic voting; multi-View learning

Domaines: Apprentissage [cs.LG]

Hongliu Cao, Simon Bernard, Laurent Heutte, Robert Sabourin. Dynamic Voting in Multi-view Learning for Radiomics Applications. IAPR Joint International Workshops on Statistical Techniques in Pattern Recognition (SPR 2018) and Structural and Syntactic Pattern Recogntion (SSPR 2018), S+SSPR, Aug 2018, Beijing, China. pp.32-41, ⟨10.1007/978-3-319-97785-0_4⟩. ⟨hal-02088181⟩