A functional model for classifying metastatic lesions integrating scans and biomarkers

Yuan Wang, Jianhua Hu, Chaan S Ng, Brian Paul Hobbs

Research output: Contribution to journalArticle

Abstract

Perfusion computed tomography is an emerging functional imaging modality that uses physiological models to quantify characteristics pertaining to the passage of fluid through blood vessels. Perfusion characteristics provide physiological correlates for neovascularization induced by tumor angiogenesis and thus a quantitative basis for cancer detection, prognostication, and treatment monitoring. We consider a liver cancer study where patients underwent a dynamic computed tomography protocol to enable evaluation of multiple perfusion characteristics derived from interrogating the time-attenuation of the concentration of the intravenously administered contrast medium. The objective is to determine the effectiveness of using perfusion characteristics to identify and discriminate between regions of liver that contain malignant tissues from normal tissue. Each patient contributes multiple regions of interest which are spatially correlated due to the shared vasculature. We propose a multivariate functional data model to disclose the correlation over time and space as well as the correlation among multiple perfusion characteristics. We further propose a simultaneous classification approach that utilizes all the correlation information to predict class assignments for collections of regions. The proposed method outperforms conventional classification approaches in the presence of strong spatial correlation. The method offers maximal relative improvement in the presence of temporal sparsity wherein measurements are obtainable at only a few time points.

Original languageEnglish (US)
JournalStatistical Methods in Medical Research
DOIs
StatePublished - Jan 1 2019

Fingerprint

Functional Model
Biomarkers
Perfusion
Computed Tomography
Liver
Cancer
Physiologic Neovascularization
Tomography
Physiological Model
Multiple Correlation
Angiogenesis
Functional Data
Blood Vessels
Multivariate Data
Spatial Correlation
Liver Neoplasms
Region of Interest
Sparsity
Attenuation
Correlate

Keywords

  • Computed tomography perfusion
  • correlated data
  • kernel smoothing
  • simultaneous decision
  • spatial and temporal correlation

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability
  • Health Information Management

Cite this

A functional model for classifying metastatic lesions integrating scans and biomarkers. / Wang, Yuan; Hu, Jianhua; Ng, Chaan S; Hobbs, Brian Paul.

In: Statistical Methods in Medical Research, 01.01.2019.

Research output: Contribution to journalArticle

@article{5acb7220e2844234877901207b43f0fe,
title = "A functional model for classifying metastatic lesions integrating scans and biomarkers",
abstract = "Perfusion computed tomography is an emerging functional imaging modality that uses physiological models to quantify characteristics pertaining to the passage of fluid through blood vessels. Perfusion characteristics provide physiological correlates for neovascularization induced by tumor angiogenesis and thus a quantitative basis for cancer detection, prognostication, and treatment monitoring. We consider a liver cancer study where patients underwent a dynamic computed tomography protocol to enable evaluation of multiple perfusion characteristics derived from interrogating the time-attenuation of the concentration of the intravenously administered contrast medium. The objective is to determine the effectiveness of using perfusion characteristics to identify and discriminate between regions of liver that contain malignant tissues from normal tissue. Each patient contributes multiple regions of interest which are spatially correlated due to the shared vasculature. We propose a multivariate functional data model to disclose the correlation over time and space as well as the correlation among multiple perfusion characteristics. We further propose a simultaneous classification approach that utilizes all the correlation information to predict class assignments for collections of regions. The proposed method outperforms conventional classification approaches in the presence of strong spatial correlation. The method offers maximal relative improvement in the presence of temporal sparsity wherein measurements are obtainable at only a few time points.",
keywords = "Computed tomography perfusion, correlated data, kernel smoothing, simultaneous decision, spatial and temporal correlation",
author = "Yuan Wang and Jianhua Hu and Ng, {Chaan S} and Hobbs, {Brian Paul}",
year = "2019",
month = "1",
day = "1",
doi = "10.1177/0962280218823795",
language = "English (US)",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",

}

TY - JOUR

T1 - A functional model for classifying metastatic lesions integrating scans and biomarkers

AU - Wang, Yuan

AU - Hu, Jianhua

AU - Ng, Chaan S

AU - Hobbs, Brian Paul

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Perfusion computed tomography is an emerging functional imaging modality that uses physiological models to quantify characteristics pertaining to the passage of fluid through blood vessels. Perfusion characteristics provide physiological correlates for neovascularization induced by tumor angiogenesis and thus a quantitative basis for cancer detection, prognostication, and treatment monitoring. We consider a liver cancer study where patients underwent a dynamic computed tomography protocol to enable evaluation of multiple perfusion characteristics derived from interrogating the time-attenuation of the concentration of the intravenously administered contrast medium. The objective is to determine the effectiveness of using perfusion characteristics to identify and discriminate between regions of liver that contain malignant tissues from normal tissue. Each patient contributes multiple regions of interest which are spatially correlated due to the shared vasculature. We propose a multivariate functional data model to disclose the correlation over time and space as well as the correlation among multiple perfusion characteristics. We further propose a simultaneous classification approach that utilizes all the correlation information to predict class assignments for collections of regions. The proposed method outperforms conventional classification approaches in the presence of strong spatial correlation. The method offers maximal relative improvement in the presence of temporal sparsity wherein measurements are obtainable at only a few time points.

AB - Perfusion computed tomography is an emerging functional imaging modality that uses physiological models to quantify characteristics pertaining to the passage of fluid through blood vessels. Perfusion characteristics provide physiological correlates for neovascularization induced by tumor angiogenesis and thus a quantitative basis for cancer detection, prognostication, and treatment monitoring. We consider a liver cancer study where patients underwent a dynamic computed tomography protocol to enable evaluation of multiple perfusion characteristics derived from interrogating the time-attenuation of the concentration of the intravenously administered contrast medium. The objective is to determine the effectiveness of using perfusion characteristics to identify and discriminate between regions of liver that contain malignant tissues from normal tissue. Each patient contributes multiple regions of interest which are spatially correlated due to the shared vasculature. We propose a multivariate functional data model to disclose the correlation over time and space as well as the correlation among multiple perfusion characteristics. We further propose a simultaneous classification approach that utilizes all the correlation information to predict class assignments for collections of regions. The proposed method outperforms conventional classification approaches in the presence of strong spatial correlation. The method offers maximal relative improvement in the presence of temporal sparsity wherein measurements are obtainable at only a few time points.

KW - Computed tomography perfusion

KW - correlated data

KW - kernel smoothing

KW - simultaneous decision

KW - spatial and temporal correlation

UR - http://www.scopus.com/inward/record.url?scp=85061057398&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061057398&partnerID=8YFLogxK

U2 - 10.1177/0962280218823795

DO - 10.1177/0962280218823795

M3 - Article

JO - Statistical Methods in Medical Research

T2 - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

ER -