A deep matrix factorization framework for identifying underlying tissue-specific patterns of DCE-MRI: applications for molecular subtype classification in breast cancer

Ming Fan, Wei Yuan, Weifen Liu, Xin Gao, Maosheng Xu, Shiwei Wang, Lihua Li

Research output: Contribution to journalArticlepeer-review


Objective Breast cancer is heterogeneous in that different angiogenesis and blood flow characteristics could be present within a tumor. The pixel kinetics of DCE-MRI can assume several distinct signal patterns related to specific tissue characteristics. Identification of the latent, tissue-specific dynamic patterns of intratumor heterogeneity can shed light on the biological mechanisms underlying the heterogeneity of tumors. Approach To mine this information, we propose a deep matrix factorization-based dynamic decomposition (DMFDE) model specifically designed according to DCE-MRI characteristics. The time-series imaging data were decomposed into tissue-specific dynamic patterns and their corresponding proportion maps. The image pixel matrix and the reference matrix of population-level kinetics obtained by clustering the dynamic signals were used as the inputs. Two multilayer neural network branches were designed to collaboratively project the input matrix into a latent dynamic pattern and a dynamic proportion matrix, which was justified using simulated data. Clinical implications of DMFDE were assessed by radiomics analysis of proportion maps obtained from the tumor/parenchyma region for classifying the luminal A subtype. Main results The decomposition performance of DMFDE was evaluated by the root mean square error (RMSE) and was shown to be better than that of the conventional convex analysis of mixtures (CAM) method. The predictive model with K=3, 4, and 5 dynamic proportion maps generated AUC values of 0.780, 0.786 and 0.790, respectively, in distinguishing between luminal A and nonluminal A tumors, which are better than the CAM method (AUC=0.726). The combination of statistical features from images with different proportion maps has the highest prediction value (AUC= 0.813), which is significantly higher than that based on CAM. Conclusion This proposed method identified the latent dynamic patterns associated with different molecular subtypes, and radiomics analysis based on the pixel compositions of the uncovered dynamic patterns was able to determine molecular subtypes of breast cancer.
Original languageEnglish (US)
JournalPhysics in Medicine & Biology
StatePublished - Nov 17 2021

Bibliographical note

KAUST Repository Item: Exported on 2021-11-24
Acknowledged KAUST grant number(s): OSR, REI/1/0018-01-01, REI/1/4216-01-01, URF/1/4352-01-01
Acknowledgements: This work was supported by the National Key R&D Program of China (2021YFE0203700), National Natural Science Foundation of China (61871428, 61731008), the Natural Science Foundation of Zhejiang Province of China (LJ19H180001), and by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under award nos. REI/1/0018-01-01, REI/1/4216-01-01, REI/1/4216-01-01, and URF/1/4352-01-01.

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology


Dive into the research topics of 'A deep matrix factorization framework for identifying underlying tissue-specific patterns of DCE-MRI: applications for molecular subtype classification in breast cancer'. Together they form a unique fingerprint.

Cite this