DI-UMONS : Dépôt institutionnel de l’université de Mons

Recherche transversale
(titres de publication, de périodique et noms de colloque inclus)
2018-03-03 - Colloque/Article dans les actes avec comité de lecture - Anglais - 17 page(s)

El Adoui Mohammed , Drisis Stylianos, Benjelloun Mohammed , "A PRM approach for early prediction of breast cancer response to chemotherapy based on registered MR images" in Computer Assisted Radiology and Surgery, Berlin , Germany, 2018

  • Codes CREF : Techniques d'imagerie et traitement d'images (DI2770)
  • Unités de recherche UMONS : Informatique (F114)
  • Instituts UMONS : Institut de Recherche en Technologies de l’Information et Sciences de l’Informatique (InforTech)
Texte intégral :

Abstract(s) :

(Anglais) Purpose: This study aims to optimize and provide a performing algorithm predicting the breast cancer response rate to the first chemotherapy using Magnetic Resonance Imaging (MRI), to early recognize the breast tumor reaction to chemotherapy, by using the Parametric Response Map (PRM) method. Methods: PRM predict the breast cancer response to chemotherapy by analyzing voxel by voxel the temporal intra tumor changes. Indeed, the tumor recognizes intra-tumor changes concerning the tumor vascularity, which is an important criterion in the present study. This method is mainly based on a spatial image affine registration between breast tumor MRI volumes before and after the first chemotherapy and a region growing segmentation of tumor volume. To evaluate our method, we used a retrospective study of 40 patients provided by our collaborating institute. Results: PRM allows to offer a color map with the percentages of positive response, negative response and stability of breast tumor response during the first chemotherapy, with identifying each region with its response rate. We assess the accuracy of the proposed method using a ground truth results based on a manual method, and then, we calculated the accuracy of the PRM based on the standard reference of the anapathologie. The AUC (Area Under the Curve) of the ROC curve of the proposed method was 0.97. Conclusion: We performed and evaluated the PRM method, that we propose to study and analyze the behavior of a tumor during the first chemotherapy, based on the intra tumor changes of MR images of breast tumor. The AUC obtained for the PRM method is considered as relevant to the early prediction of breast tumor response.

Notes :
  • (Anglais) This article has been accepted by the CARS 2018 conference as a talk presentation in Berlin, Germany (June 20–23, 2018)