Advanced machine learning informatics modeling using clinical and radiological imaging metrics for characterizing breast tumor characteristics with the oncotypedx gene array.

Research output: Contribution to journalArticlepeer-review

Abstract

Purpose: Emerging data on breast cancer suggest that different breast cancer phenotypes may respond differently to available adjuvant therapies. Optimal use of established and imaging methods, such as multiparametric magnetic resonance imaging (mpMRI) can simultaneously identify key functional parameters and provide unique imaging phenotypes of breast cancer. Therefore, we have developed and implemented a new machine-learning informatic system that integrates clinical variables, derived from imaging and clinical health records, to compare with the 21-gene array assay, OncotypeDX. Materials and methods: We tested our informatics modeling in a subset of patients (n=81) who had ER+ disease and underwent OncotypeDX gene expression and breast mpMRI testing. The machine-learning informatic method is termed Integrated Radiomic Informatic System (IRIS) was applied to the mpMRI, clinical and pathologic descriptors, as well as a gene array analysis. The IRIS method using an advanced graph theoretic model that transforms the patient space into a visualization heatmap and quantitative metrics. Summary statistics (mean and standard deviations) for the quantitative imaging parameters were obtained. Sensitivity and specificity and Area Under the Curve were calculated for the classification of the patients. Statistical significance was set at p ≤ 0.05. Results: The OncotypeDX classification by IRIS model had sensitivity of 95% and specificity of 89% with AUC of 0.92 with 19(low), 50 (intermediate), 12(high) patients classified in each risk group. The breast lesion size was larger for the high-risk group (7.6±5.8cm2) and lower for both low risk (5.8±9.0cm2) and intermediate risk (4.6±5.4cm2) groups. There were significant differences in PK-DCE and ADC map values in each group. The lesion ADC map values for high- and intermediate-risk groups were significantly lower than the low-risk group (1.09 vs 1.38x10-3mm2/s). Conclusion: These initial studies provide deeper understandings of imaging features and molecular gene array OncotypeDX score. This insight provides the foundation to relate these imaging features to the assessment of treatment response for improved personalized medicine.

Original languageEnglish (US)
JournalUnknown Journal
StatePublished - Nov 7 2018

Keywords

  • ADC map
  • Breast
  • Cancer
  • Diffusion-weighted imaging
  • DWI
  • Informatics
  • Machine learning
  • Magnetic Resonance Imaging
  • Radiomics
  • Treatment response

ASJC Scopus subject areas

  • General

Fingerprint Dive into the research topics of 'Advanced machine learning informatics modeling using clinical and radiological imaging metrics for characterizing breast tumor characteristics with the oncotypedx gene array.'. Together they form a unique fingerprint.

Cite this