Treatment outcome prediction for cancer patients based on radiomics and belief function theory

Jian Wu, Chunfeng Lian, Su Ruan, Thomas R. Mazur, Sasa Mutic, Mark A. Anastasio, Perry W. Grigsby, Pierre Vera, Hua Li

Research output: Contribution to journalArticlepeer-review

17 Scopus citations

Abstract

In this paper, we proposed a new radiomics-based treatment outcome prediction model for cancer patients. The prediction model is developed based on belief function theory and sparsity learning to address the challenges of redundancy, heterogeneity, and uncertainty of radiomic features, and relatively small-sized and unbalanced training samples. The model first selects the most predictive feature subsets from relatively large amounts of radiomic features extracted from pre- and/or in-treatment positron emission tomography images and available clinical and demographic features. Then an evidential k-nearest neighbor classifier is proposed to utilize the selected features for treatment outcome prediction. Twenty-five stage II-III lung, 36 esophagus, 63 stage II-III cervix, and 45 lymphoma cancer patient cases were included in this retrospective study. Performance and robustness of the proposed model were assessed with measures of feature selection stability, outcome prediction accuracy, and receiver operating characteristics analysis. Comparison with other methods was conducted to demonstrate the feasibility and superior performance of the proposed model.

Original languageEnglish
Article number8474291
Pages (from-to)216-224
Number of pages9
JournalIEEE Transactions on Radiation and Plasma Medical Sciences
Volume3
Issue number2
DOIs
StatePublished - Mar 2019

Keywords

  • Belief function theory (BFT)
  • cancer therapy
  • positron emission tomography (PET) images
  • radiomics
  • treatment outcome prediction

Fingerprint

Dive into the research topics of 'Treatment outcome prediction for cancer patients based on radiomics and belief function theory'. Together they form a unique fingerprint.

Cite this