Robust automatic breast cancer staging using a combination of functional genomics and image-omics

Abstract

Breast cancer is one of the leading cancers worldwide. Precision medicine is a new trend that systematically examines molecular and functional genomic information within each patient's cancer to identify the patterns that may affect treatment decisions and potential outcomes. As a part of precision medicine, computer-aided diagnosis enables joint analysis of functional genomic information and image from pathological images. In this paper we propose an integrated framework for breast cancer staging using image-omics and functional genomic information. The entire biomedical imaging informatics framework consists of image-omics extraction, feature combination, and classification. First, a robust automatic nuclei detection and segmentation is presented to identify tumor regions, delineate nuclei boundaries and calculate a set of image-based morphological features; next, the low dimensional image-omics is obtained through principal component analysis and is concatenated with the functional genomic features identified by a linear model. A support vector machine for differentiating stage I breast cancer from other stages are learned. We experimentally demonstrate that compared with a single type of representation (image-omics), the combination of image-omics and functional genomic feature can improve the classification accuracy by 3%.

Figures

Fig. 1

Bargraph of gene expression values for gene C19orf33 and SLC4A8 discovered from linear modeling with breast tumor stages.

Fig. 2

Misclassification rates with respect to different parameters using Gaussian kernel to differentiate the patients at stage I and the other stages. (a) The misclassification rates based on image-omics only; (b) The misclassification rate based on image-omics and the two identified functional genomic features.

Fig. 3

The receiver operation curves of the classification with image-omics only and a combination of the morphological features and the two identified functional genomic features.

Fig. 4

(a) A randomly picked original image; (b) The nuclei detection result; (c) The tumor region (marked in light green) segmentation result; (d) The nuclei identified for feature extraction.