期刊
IEEE TRANSACTIONS ON MEDICAL IMAGING
卷 35, 期 1, 页码 76-88出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TMI.2015.2456188
关键词
Terms-Quantitative histomorphometry; digital pathology; dimensionality reduction; feature selection
类别
资金
- National Cancer Institute of the National Institutes of Health [R01CA136535-013, R01CA140772-01, R21CA167811-01, R21CA179327-01, R21CA195152-01]
- National Institute of Diabetes and Digestive and Kidney Diseases [R01DK098503-02]
- DOD Prostate Cancer Synergistic Idea Development Award [PC120857]
- DOD Lung Cancer Idea Development New Investigator Award [LC130463]
- DOD Prostate Cancer Idea Development Award
- Ohio Third Frontier Technology development Grant
- CTSC Coulter Annual Pilot Grant
- Case Comprehensive Cancer Center Pilot Grant
- Cleveland Clinic
- Wallace H. Coulter Foundation Program in the Department of Biomedical Engineering at Case Western Reserve University
- National Science Foundation Graduate Research Fellowship Program
- NATIONAL CANCER INSTITUTE [R21CA167811, R21CA179327, R01CA140772, R21CA195152, R01CA136535] Funding Source: NIH RePORTER
- NATIONAL INSTITUTE OF DIABETES AND DIGESTIVE AND KIDNEY DISEASES [R01DK098503] Funding Source: NIH RePORTER
Quantitative histomorphometry (QH) refers to the process of computationally modeling disease appearance on digital pathology images by extracting hundreds of image features and using them to predict disease presence or outcome. Since constructing a robust and interpretable classifier is challenging in a high dimensional feature space, dimensionality reduction (DR) is often implemented prior to classifier construction. However, when DR is performed it can be challenging to quantify the contribution of each of the original features to the final classification result. We have previously presented a method for scoring features based on their importance for classification on an embedding derived via principal components analysis (PCA). However, nonlinear DR involves the eigen-decomposition of a kernel matrix rather than the data itself, compounding the issue of classifier interpretability. In this paper we present feature importance in nonlinear embeddings (FINE), an extension of our PCA-based feature scoring method to kernel PCA (KPCA), as well as several NLDR algorithms that can be cast as variants of KPCA. FINE is applied to four digital pathology datasets to identify key QH features for predicting the risk of breast and prostate cancer recurrence. Measures of nuclear and glandular architecture and clusteredness were found to play an important role in predicting the likelihood of recurrence of both breast and prostate cancers. Compared to the t-test, Fisher score, and Gini index, FINE was able to identify a stable set of features that provide good classification accuracy on four publicly available datasets from the NIPS 2003 Feature Selection Challenge.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据