Cancer screening with early detection and intervention is a key factor for successful treatment and has shown potential to improve survival and reduce mortality of the cancer patients. ROC based methods have been extensively utilized to evaluate the performance of early detection. However, for evaluating early detection tests of cancer, ROC curves have been either underused or not always used in the best manner. Cancer researchers also need efficient methods to identify risk factors (biomarkers) that could improve the performance of early detection. Therefore, it is crucial to develop new methods for identifying and evaluating the prognostic and diagnostic risk factors (biomarkers) for earlier cancer detection and risk assessment. This pilot project focuses on developing novel ROC based computational models for risk factor identification, biomarker evaluation, and cancer prediction in cancer screening. Both linear and nonlinear models will be developed. It will provide efficient computational tools for earlier cancer detection, prevention, and risk assessment.

Public Health Relevance

Relevant Statement This research will provide computational tools and software for early cancer detection and intervention. It will benefit the whole cancer research society. Upon completing the pilot project, investigators will have the more efficient tools for biomarker discovery and performance evaluation

Agency
National Institute of Health (NIH)
Institute
National Cancer Institute (NCI)
Type
Small Research Grants (R03)
Project #
5R03CA133899-02
Application #
7939576
Study Section
Special Emphasis Panel (ZCA1-SRLB-F (O1))
Program Officer
Krueger, Karl E
Project Start
2009-09-28
Project End
2013-08-31
Budget Start
2010-09-01
Budget End
2013-08-31
Support Year
2
Fiscal Year
2010
Total Cost
$75,000
Indirect Cost
Name
University of Maryland Baltimore
Department
Public Health & Prev Medicine
Type
Schools of Medicine
DUNS #
188435911
City
Baltimore
State
MD
Country
United States
Zip Code
21201
Liu, Zhenqiu; Chen, Dechang; Sheng, Li et al. (2013) Class prediction and feature selection with linear optimization for metagenomic count data. PLoS One 8:e53253
Liu, Zhenqiu; Bensmail, Halima; Tan, Ming (2012) Efficient feature selection and multiclass classification with integrated instance and model based learning. Evol Bioinform Online 8:197-205
Liu, Zhenqiu; Hsiao, William; Cantarel, Brandi L et al. (2011) Sparse distance-based learning for simultaneous multiclass classification and feature selection of metagenomic data. Bioinformatics 27:3242-9
Liu, Zhenqiu; Chen, Dechang; Tan, Ming et al. (2010) Kernel based methods for accelerated failure time model with ultra-high dimensional data. BMC Bioinformatics 11:606
Liu, Zhenqiu; Chen, Dechang; Tian, Guoliang et al. (2010) Efficient support vector machine method for survival prediction with SEER data. Adv Exp Med Biol 680:11-8