1: \begin{abstract}

2: \singlespacing

3: 

4: 

5: This work  is motivated by multimodality breast cancer imaging data, which is quite challenging  in that the signals of discrete  tumor-associated microvesicles (TMVs) are randomly distributed  with  heterogeneous patterns.  This  imposes a significant challenge  for conventional imaging regression and dimension reduction models assuming a homogeneous feature structure.

6: We develop an innovative multilayer tensor learning method to  incorporate heterogeneity to a higher-order tensor decomposition and predict disease status effectively through utilizing subject-wise imaging features and multimodality information.  Specifically, we construct a multilayer decomposition which leverages an individualized imaging layer in addition to a modality-specific  tensor structure.

7: One major advantage of our approach is that we are able to efficiently  capture the heterogeneous  spatial features of signals that are not characterized by a population structure as well as integrating  multimodality  information simultaneously.

8: To achieve scalable computing, we develop a new bi-level block improvement algorithm. In theory, we investigate both the algorithm convergence property, tensor signal recovery error bound and   asymptotic consistency for prediction model estimation. We also apply the proposed method for  simulated  and human breast cancer imaging data. Numerical results demonstrate that the proposed method outperforms other existing competing methods.

9: 

10: 

11: \bigskip

12: 

13: \noindent \textbf {Key words:}  Cancer imaging, dimension reduction, heterogeneous modeling, high-order tensor decomposition, multimodality integration, spatial information.

14: 

15: \end{abstract}

16: