Prostate Cancer Detection on Dynamic Contrast-Enhanced MRI: Computer-Aided Diagnosis Versus Single Perfusion Parameter Maps

Sung, Yu Sub; Kwon, Heon‐Ju; Park, Bum-Woo; Cho, Gyunggoo; Lee, Chang Kyung; Cho, Kyoung‐Sik; Kim, Jeong Kon

doi:10.2214/ajr.10.6062

Cited by 68 publications

(62 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is expected since both use the Metropolis-Hastings Markov chain Monte Carlo (MCMC) method and similar prior information. Their only difference is that we parameterized the posterior probability distribution function p(k,σ|CTIC) with ve to optimize EES volume directly instead of calculating it via kep (ve=Ktrans/kep) [7].…”

Section: Discussionmentioning

confidence: 99%

“…Tracer kinetic models such as the extended Toft model [5] that describe the enhancement process are often used to derive quantitative parameters and are increasingly used in diagnostic models [6] including computer aided diagnostic (CAD) software [7,8]. Accurate quantification that will be reproducible between different clinical sites is necessary for the widespread of DCE based CAD software.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A comparison of Bayesian and non-linear regression methods for robust estimation of pharmacokinetics in DCE-MRI and how it affects cancer diagnosis

Δικαίος

Atkinson

Tudisca

et al. 2017

Computerized Medical Imaging and Graphics

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A comparison of Bayesian and non-linear regression methods for robust estimation of pharmacokinetics in DCE-MRI and how it affects cancer diagnosis

Δικαίος

Atkinson

Tudisca

et al. 2017

Computerized Medical Imaging and Graphics

View full text Add to dashboard Cite

show abstract

“…In the entire prostate mass , accuracy was greater for CAD than for all perfusion parameters or T2WI (68 -76%); sensitivity was higher for CAD than for T2WI, initial slope, wash-in rate, slope , and washout rate (39-78%); and specificity was greater for CAD than for T2WI, k(ep), k(el), and time to peak (57-69%) (p < 0.01). Reseach concluded CAD can improve the diagnostic performance of DCE-MRI in prostate carcinoma detection, which may vary according to zone anatomy (18) . In a research by Sayed Zidan et al in 2015 , they found sensitivity of DCE-MRI, ADC at 1.2 and ADC at 1.4 in detection of prostatic carcinoma was 99 %, 84.7% and 99% respectively (P = 0.0001).…”

Section: Quantitative Approachmentioning

confidence: 99%

MRI in Diagnosis of Prostate Carcinoma

MBBS¹

2017

jmscr

View full text Add to dashboard Cite

“…Recent studies showed an increasing interest in developing automatic diagnosis systems to detect and characterize prostate cancer on the basis of a multi-parametric MR imaging approach (Vos et al 2012, Niaf et al 2012, Sung et al 2011. The aim was to assist radiologists in making correct diagnosis decisions by providing an objective and reproducible malignancy score for suspicious targets (Niaf et al 2012, Shah et al 2012.…”

Section: Introductionmentioning

confidence: 99%

Comparison between PUN and Tofts models in the quantification of dynamic contrast-enhanced MR imaging

Mazzetti¹,

Gliozzi²,

Bracco³

et al. 2012

Phys. Med. Biol.

View full text Add to dashboard Cite

Dynamic contrast-enhanced study in magnetic resonance imaging (DCE-MRI) is an important tool in oncology to visualize tissues vascularization and to define tumour aggressiveness on the basis of an altered perfusion and permeability. Pharmacokinetic models are generally used to extract hemodynamic parameters, providing a quantitative description of the contrast uptake and wash-out. Empirical functions can also be used to fit experimental data without the need of any assumption about tumour physiology, as in pharmacokinetic models, increasing their diagnostic utility, in particular when automatic diagnosis systems are implemented on the basis of an MRI multiparametric approach. Phenomenological universalities (PUN) represent a novel tool for experimental research and offer a simple and systematic method to represent a set of data independent of the application field. DCE-MRI acquisitions can thus be advantageously evaluated by the extended PUN class, providing a convenient diagnostic tool to analyse functional studies, adding a new set of features for the classification of malignant and benign lesions in computer aided detection systems. In this work the Tofts pharmacokinetic model and the class EU1 generated by the PUN description were compared in the study of DCE-MRI of the prostate, evaluating complexity of model implementation, goodness of fitting results, classification performances and computational cost. The mean R 2 obtained with the EU1 and Tofts model were equal to 0.96 and 0.90, respectively, and the classification performances achieved by the EU1 model and the Tofts implementation discriminated malignant from benign tissues with an area under the receiver operating characteristic curve equal to 0.92 and 0.91, respectively. Furthermore, the EU1 model has a simpler functional form which reduces implementation complexity

show abstract

Prostate Cancer Detection on Dynamic Contrast-Enhanced MRI: Computer-Aided Diagnosis Versus Single Perfusion Parameter Maps

Abstract: CAD can improve the diagnostic performance of DCE-MRI in prostate cancer detection, which may vary according to zonal anatomy.

Cited by 68 publications

References 34 publications

A comparison of Bayesian and non-linear regression methods for robust estimation of pharmacokinetics in DCE-MRI and how it affects cancer diagnosis

A comparison of Bayesian and non-linear regression methods for robust estimation of pharmacokinetics in DCE-MRI and how it affects cancer diagnosis

MRI in Diagnosis of Prostate Carcinoma

Comparison between PUN and Tofts models in the quantification of dynamic contrast-enhanced MR imaging

Contact Info

Product

Resources

About