Fractal-based radiomic approach to predict complete pathological response after chemo-radiotherapy in rectal cancer

Davide Cusumano, Nicola Dinapoli, Luca Boldrini, Giuditta Chiloiro, Roberto Gatta, Carlotta Masciocchi, Jacopo Lenkowicz, Calogero Casà, Andrea Damiani, Luigi Azario, Johan Van Soest, Andre Dekker, Philippe Lambin, Marco De Spirito, Vincenzo Valentini

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

The aim of this study was to propose a methodology to investigate the tumour heterogeneity and evaluate its ability to predict pathologically complete response (pCR) after chemo-radiotherapy (CRT) in locally advanced rectal cancer (LARC). This approach consisted in normalising the pixel intensities of the tumour and identifying the different sub-regions using an intensity-based thresholding. The spatial organisation of these subpopulations was quantified using the fractal dimension (FD). This approach was implemented in a radiomic workflow and applied to 198 T2-weighted pre-treatment magnetic resonance (MR) images of LARC patients. Three types of features were extracted from the gross tumour volume (GTV): morphological, statistical and fractal features. Feature selection was performed using the Wilcoxon test and a logistic regression model was calculated to predict the pCR probability after CRT. The model was elaborated considering the patients treated in two institutions: Fondazione Policlinico Universitario “Agostino Gemelli” of Rome (173 cases, training set) and University Medical Centre of Maastricht (25 cases, validation set). The results obtained showed that the fractal parameters of the subpopulations have the highest performance in predicting pCR. The predictive model elaborated had an area under the curve (AUC) equal to 0.77 ± 0.07. The model reliability was confirmed by the validation set (AUC = 0.79 ± 0.09). This study suggests that the fractal analysis can play an important role in radiomics, providing valuable information not only about the GTV structure, but also about its inner subpopulations.

Original languageEnglish
Pages (from-to)286-295
Number of pages10
JournalRadiologia Medica
Volume123
Issue number4
DOIs
Publication statusPublished - Apr 1 2018

Fingerprint

Fractals
Rectal Neoplasms
Radiotherapy
Tumor Burden
Area Under Curve
Logistic Models
Workflow
Neoplasms
Magnetic Resonance Spectroscopy
Therapeutics

Keywords

  • Fractals
  • Magnetic resonance imaging
  • Predictive model
  • Radiomics
  • Rectal cancer

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Fractal-based radiomic approach to predict complete pathological response after chemo-radiotherapy in rectal cancer. / Cusumano, Davide; Dinapoli, Nicola; Boldrini, Luca; Chiloiro, Giuditta; Gatta, Roberto; Masciocchi, Carlotta; Lenkowicz, Jacopo; Casà, Calogero; Damiani, Andrea; Azario, Luigi; Van Soest, Johan; Dekker, Andre; Lambin, Philippe; De Spirito, Marco; Valentini, Vincenzo.

In: Radiologia Medica, Vol. 123, No. 4, 01.04.2018, p. 286-295.

Research output: Contribution to journalArticle

Cusumano, D, Dinapoli, N, Boldrini, L, Chiloiro, G, Gatta, R, Masciocchi, C, Lenkowicz, J, Casà, C, Damiani, A, Azario, L, Van Soest, J, Dekker, A, Lambin, P, De Spirito, M & Valentini, V 2018, 'Fractal-based radiomic approach to predict complete pathological response after chemo-radiotherapy in rectal cancer', Radiologia Medica, vol. 123, no. 4, pp. 286-295. https://doi.org/10.1007/s11547-017-0838-3
Cusumano, Davide ; Dinapoli, Nicola ; Boldrini, Luca ; Chiloiro, Giuditta ; Gatta, Roberto ; Masciocchi, Carlotta ; Lenkowicz, Jacopo ; Casà, Calogero ; Damiani, Andrea ; Azario, Luigi ; Van Soest, Johan ; Dekker, Andre ; Lambin, Philippe ; De Spirito, Marco ; Valentini, Vincenzo. / Fractal-based radiomic approach to predict complete pathological response after chemo-radiotherapy in rectal cancer. In: Radiologia Medica. 2018 ; Vol. 123, No. 4. pp. 286-295.
@article{8b69d51a3a1545e39ff334afad4dcdab,
title = "Fractal-based radiomic approach to predict complete pathological response after chemo-radiotherapy in rectal cancer",
abstract = "The aim of this study was to propose a methodology to investigate the tumour heterogeneity and evaluate its ability to predict pathologically complete response (pCR) after chemo-radiotherapy (CRT) in locally advanced rectal cancer (LARC). This approach consisted in normalising the pixel intensities of the tumour and identifying the different sub-regions using an intensity-based thresholding. The spatial organisation of these subpopulations was quantified using the fractal dimension (FD). This approach was implemented in a radiomic workflow and applied to 198 T2-weighted pre-treatment magnetic resonance (MR) images of LARC patients. Three types of features were extracted from the gross tumour volume (GTV): morphological, statistical and fractal features. Feature selection was performed using the Wilcoxon test and a logistic regression model was calculated to predict the pCR probability after CRT. The model was elaborated considering the patients treated in two institutions: Fondazione Policlinico Universitario “Agostino Gemelli” of Rome (173 cases, training set) and University Medical Centre of Maastricht (25 cases, validation set). The results obtained showed that the fractal parameters of the subpopulations have the highest performance in predicting pCR. The predictive model elaborated had an area under the curve (AUC) equal to 0.77 ± 0.07. The model reliability was confirmed by the validation set (AUC = 0.79 ± 0.09). This study suggests that the fractal analysis can play an important role in radiomics, providing valuable information not only about the GTV structure, but also about its inner subpopulations.",
keywords = "Fractals, Magnetic resonance imaging, Predictive model, Radiomics, Rectal cancer",
author = "Davide Cusumano and Nicola Dinapoli and Luca Boldrini and Giuditta Chiloiro and Roberto Gatta and Carlotta Masciocchi and Jacopo Lenkowicz and Calogero Cas{\`a} and Andrea Damiani and Luigi Azario and {Van Soest}, Johan and Andre Dekker and Philippe Lambin and {De Spirito}, Marco and Vincenzo Valentini",
year = "2018",
month = "4",
day = "1",
doi = "10.1007/s11547-017-0838-3",
language = "English",
volume = "123",
pages = "286--295",
journal = "Radiologia Medica",
issn = "0033-8362",
publisher = "Springer-Verlag Italia s.r.l.",
number = "4",

}

TY - JOUR

T1 - Fractal-based radiomic approach to predict complete pathological response after chemo-radiotherapy in rectal cancer

AU - Cusumano, Davide

AU - Dinapoli, Nicola

AU - Boldrini, Luca

AU - Chiloiro, Giuditta

AU - Gatta, Roberto

AU - Masciocchi, Carlotta

AU - Lenkowicz, Jacopo

AU - Casà, Calogero

AU - Damiani, Andrea

AU - Azario, Luigi

AU - Van Soest, Johan

AU - Dekker, Andre

AU - Lambin, Philippe

AU - De Spirito, Marco

AU - Valentini, Vincenzo

PY - 2018/4/1

Y1 - 2018/4/1

N2 - The aim of this study was to propose a methodology to investigate the tumour heterogeneity and evaluate its ability to predict pathologically complete response (pCR) after chemo-radiotherapy (CRT) in locally advanced rectal cancer (LARC). This approach consisted in normalising the pixel intensities of the tumour and identifying the different sub-regions using an intensity-based thresholding. The spatial organisation of these subpopulations was quantified using the fractal dimension (FD). This approach was implemented in a radiomic workflow and applied to 198 T2-weighted pre-treatment magnetic resonance (MR) images of LARC patients. Three types of features were extracted from the gross tumour volume (GTV): morphological, statistical and fractal features. Feature selection was performed using the Wilcoxon test and a logistic regression model was calculated to predict the pCR probability after CRT. The model was elaborated considering the patients treated in two institutions: Fondazione Policlinico Universitario “Agostino Gemelli” of Rome (173 cases, training set) and University Medical Centre of Maastricht (25 cases, validation set). The results obtained showed that the fractal parameters of the subpopulations have the highest performance in predicting pCR. The predictive model elaborated had an area under the curve (AUC) equal to 0.77 ± 0.07. The model reliability was confirmed by the validation set (AUC = 0.79 ± 0.09). This study suggests that the fractal analysis can play an important role in radiomics, providing valuable information not only about the GTV structure, but also about its inner subpopulations.

AB - The aim of this study was to propose a methodology to investigate the tumour heterogeneity and evaluate its ability to predict pathologically complete response (pCR) after chemo-radiotherapy (CRT) in locally advanced rectal cancer (LARC). This approach consisted in normalising the pixel intensities of the tumour and identifying the different sub-regions using an intensity-based thresholding. The spatial organisation of these subpopulations was quantified using the fractal dimension (FD). This approach was implemented in a radiomic workflow and applied to 198 T2-weighted pre-treatment magnetic resonance (MR) images of LARC patients. Three types of features were extracted from the gross tumour volume (GTV): morphological, statistical and fractal features. Feature selection was performed using the Wilcoxon test and a logistic regression model was calculated to predict the pCR probability after CRT. The model was elaborated considering the patients treated in two institutions: Fondazione Policlinico Universitario “Agostino Gemelli” of Rome (173 cases, training set) and University Medical Centre of Maastricht (25 cases, validation set). The results obtained showed that the fractal parameters of the subpopulations have the highest performance in predicting pCR. The predictive model elaborated had an area under the curve (AUC) equal to 0.77 ± 0.07. The model reliability was confirmed by the validation set (AUC = 0.79 ± 0.09). This study suggests that the fractal analysis can play an important role in radiomics, providing valuable information not only about the GTV structure, but also about its inner subpopulations.

KW - Fractals

KW - Magnetic resonance imaging

KW - Predictive model

KW - Radiomics

KW - Rectal cancer

UR - http://www.scopus.com/inward/record.url?scp=85037703167&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85037703167&partnerID=8YFLogxK

U2 - 10.1007/s11547-017-0838-3

DO - 10.1007/s11547-017-0838-3

M3 - Article

C2 - 29230678

AN - SCOPUS:85037703167

VL - 123

SP - 286

EP - 295

JO - Radiologia Medica

JF - Radiologia Medica

SN - 0033-8362

IS - 4

ER -