Statistical clustering of parametric maps from dynamic contrast enhanced MRI and an associated decision tree model for non-invasive tumour grading of T1b solid clear cell renal cell carcinoma

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Objectives: To apply a statistical clustering algorithm to combine information from dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) into a single tumour map to distinguish high-grade from low-grade T1b clear cell renal cell carcinoma (ccRCC). Methods: This prospective, Institutional Review Board -approved, Health Insurance Portability and Accountability Act -compliant study included 18 patients with solid T1b ccRCC who underwent pre-surgical DCE MRI. After statistical clustering of the parametric maps of the transfer constant between the intravascular and extravascular space (Ktrans), rate constant (Kep) and initial area under the concentration curve (iAUC) with a fuzzy c-means (FCM) algorithm, each tumour was segmented into three regions (low/medium/high active areas). Percentages of each region and tumour size were compared to tumour grade at histopathology. A decision-tree model was constructed to select the best parameter(s) to predict high-grade ccRCC. Results: Seven high-grade and 11 low-grade T1b ccRCCs were included. High-grade histology was associated with higher percent high active areas (p = 0.0154) and this was the only feature selected by the decision tree model, which had a diagnostic performance of 78% accuracy, 86% sensitivity, 73% specificity, 67% positive predictive value and 89% negative predictive value. Conclusions: The FCM integrates multiple DCE-derived parameter maps and identifies tumour regions with unique pharmacokinetic characteristics. Using this approach, a decision tree model using criteria beyond size to predict tumour grade in T1b ccRCCs is proposed. Key Points: • Tumour size did not correlate with tumour grade in T1b ccRCC.• Tumour heterogeneity can be analysed using statistical clustering via DCE-MRI parameters.• High-grade ccRCC has a larger percentage of high active area than low-grade ccRCCs.• A decision-tree model offers a simple way to differentiate high/low-grade ccRCCs.

Original languageEnglish (US)
Pages (from-to)1-9
Number of pages9
JournalEuropean Radiology
DOIs
StateAccepted/In press - Jul 5 2017

Fingerprint

Decision Trees
Neoplasm Grading
Renal Cell Carcinoma
Cluster Analysis
Magnetic Resonance Imaging
Neoplasms
Health Insurance Portability and Accountability Act
Research Ethics Committees
Area Under Curve
Histology
Pharmacokinetics
Sensitivity and Specificity

Keywords

  • Clear-cell renal cell carcinoma
  • Dynamic contrast-enhanced-MRI
  • Kidney cancer
  • Statistical clustering
  • Tumour heterogeneity

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

@article{066ad1ecccdd4f3f803e70039365143b,
title = "Statistical clustering of parametric maps from dynamic contrast enhanced MRI and an associated decision tree model for non-invasive tumour grading of T1b solid clear cell renal cell carcinoma",
abstract = "Objectives: To apply a statistical clustering algorithm to combine information from dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) into a single tumour map to distinguish high-grade from low-grade T1b clear cell renal cell carcinoma (ccRCC). Methods: This prospective, Institutional Review Board -approved, Health Insurance Portability and Accountability Act -compliant study included 18 patients with solid T1b ccRCC who underwent pre-surgical DCE MRI. After statistical clustering of the parametric maps of the transfer constant between the intravascular and extravascular space (Ktrans), rate constant (Kep) and initial area under the concentration curve (iAUC) with a fuzzy c-means (FCM) algorithm, each tumour was segmented into three regions (low/medium/high active areas). Percentages of each region and tumour size were compared to tumour grade at histopathology. A decision-tree model was constructed to select the best parameter(s) to predict high-grade ccRCC. Results: Seven high-grade and 11 low-grade T1b ccRCCs were included. High-grade histology was associated with higher percent high active areas (p = 0.0154) and this was the only feature selected by the decision tree model, which had a diagnostic performance of 78{\%} accuracy, 86{\%} sensitivity, 73{\%} specificity, 67{\%} positive predictive value and 89{\%} negative predictive value. Conclusions: The FCM integrates multiple DCE-derived parameter maps and identifies tumour regions with unique pharmacokinetic characteristics. Using this approach, a decision tree model using criteria beyond size to predict tumour grade in T1b ccRCCs is proposed. Key Points: • Tumour size did not correlate with tumour grade in T1b ccRCC.• Tumour heterogeneity can be analysed using statistical clustering via DCE-MRI parameters.• High-grade ccRCC has a larger percentage of high active area than low-grade ccRCCs.• A decision-tree model offers a simple way to differentiate high/low-grade ccRCCs.",
keywords = "Clear-cell renal cell carcinoma, Dynamic contrast-enhanced-MRI, Kidney cancer, Statistical clustering, Tumour heterogeneity",
author = "Yin Xi and Qing Yuan and Yue Zhang and Madhuranthakam, {Ananth J} and Michael Fulkerson and Vitaly Margulis and Brugarolas, {James B} and Payal Kapur and Cadeddu, {Jeffrey A} and Ivan Pedrosa",
year = "2017",
month = "7",
day = "5",
doi = "10.1007/s00330-017-4925-6",
language = "English (US)",
pages = "1--9",
journal = "European Radiology",
issn = "0938-7994",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Statistical clustering of parametric maps from dynamic contrast enhanced MRI and an associated decision tree model for non-invasive tumour grading of T1b solid clear cell renal cell carcinoma

AU - Xi, Yin

AU - Yuan, Qing

AU - Zhang, Yue

AU - Madhuranthakam, Ananth J

AU - Fulkerson, Michael

AU - Margulis, Vitaly

AU - Brugarolas, James B

AU - Kapur, Payal

AU - Cadeddu, Jeffrey A

AU - Pedrosa, Ivan

PY - 2017/7/5

Y1 - 2017/7/5

N2 - Objectives: To apply a statistical clustering algorithm to combine information from dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) into a single tumour map to distinguish high-grade from low-grade T1b clear cell renal cell carcinoma (ccRCC). Methods: This prospective, Institutional Review Board -approved, Health Insurance Portability and Accountability Act -compliant study included 18 patients with solid T1b ccRCC who underwent pre-surgical DCE MRI. After statistical clustering of the parametric maps of the transfer constant between the intravascular and extravascular space (Ktrans), rate constant (Kep) and initial area under the concentration curve (iAUC) with a fuzzy c-means (FCM) algorithm, each tumour was segmented into three regions (low/medium/high active areas). Percentages of each region and tumour size were compared to tumour grade at histopathology. A decision-tree model was constructed to select the best parameter(s) to predict high-grade ccRCC. Results: Seven high-grade and 11 low-grade T1b ccRCCs were included. High-grade histology was associated with higher percent high active areas (p = 0.0154) and this was the only feature selected by the decision tree model, which had a diagnostic performance of 78% accuracy, 86% sensitivity, 73% specificity, 67% positive predictive value and 89% negative predictive value. Conclusions: The FCM integrates multiple DCE-derived parameter maps and identifies tumour regions with unique pharmacokinetic characteristics. Using this approach, a decision tree model using criteria beyond size to predict tumour grade in T1b ccRCCs is proposed. Key Points: • Tumour size did not correlate with tumour grade in T1b ccRCC.• Tumour heterogeneity can be analysed using statistical clustering via DCE-MRI parameters.• High-grade ccRCC has a larger percentage of high active area than low-grade ccRCCs.• A decision-tree model offers a simple way to differentiate high/low-grade ccRCCs.

AB - Objectives: To apply a statistical clustering algorithm to combine information from dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) into a single tumour map to distinguish high-grade from low-grade T1b clear cell renal cell carcinoma (ccRCC). Methods: This prospective, Institutional Review Board -approved, Health Insurance Portability and Accountability Act -compliant study included 18 patients with solid T1b ccRCC who underwent pre-surgical DCE MRI. After statistical clustering of the parametric maps of the transfer constant between the intravascular and extravascular space (Ktrans), rate constant (Kep) and initial area under the concentration curve (iAUC) with a fuzzy c-means (FCM) algorithm, each tumour was segmented into three regions (low/medium/high active areas). Percentages of each region and tumour size were compared to tumour grade at histopathology. A decision-tree model was constructed to select the best parameter(s) to predict high-grade ccRCC. Results: Seven high-grade and 11 low-grade T1b ccRCCs were included. High-grade histology was associated with higher percent high active areas (p = 0.0154) and this was the only feature selected by the decision tree model, which had a diagnostic performance of 78% accuracy, 86% sensitivity, 73% specificity, 67% positive predictive value and 89% negative predictive value. Conclusions: The FCM integrates multiple DCE-derived parameter maps and identifies tumour regions with unique pharmacokinetic characteristics. Using this approach, a decision tree model using criteria beyond size to predict tumour grade in T1b ccRCCs is proposed. Key Points: • Tumour size did not correlate with tumour grade in T1b ccRCC.• Tumour heterogeneity can be analysed using statistical clustering via DCE-MRI parameters.• High-grade ccRCC has a larger percentage of high active area than low-grade ccRCCs.• A decision-tree model offers a simple way to differentiate high/low-grade ccRCCs.

KW - Clear-cell renal cell carcinoma

KW - Dynamic contrast-enhanced-MRI

KW - Kidney cancer

KW - Statistical clustering

KW - Tumour heterogeneity

UR - http://www.scopus.com/inward/record.url?scp=85021796190&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85021796190&partnerID=8YFLogxK

U2 - 10.1007/s00330-017-4925-6

DO - 10.1007/s00330-017-4925-6

M3 - Article

C2 - 28681074

AN - SCOPUS:85021796190

SP - 1

EP - 9

JO - European Radiology

JF - European Radiology

SN - 0938-7994

ER -