Quantitative CT based radiomics as predictor of resectability of pancreatic adenocarcinoma

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)

Abstract

In current clinical practice, the resectability of pancreatic ductal adenocarcinoma (PDA) is determined subjec-tively by a physician, which is an error-prone procedure. In this paper, we present a method for automated determination of resectability of PDA from a routine abdominal CT, to reduce such decision errors. The tumor features are extracted from a group of patients with both hypo- A nd iso-attenuating tumors, of which 29 were resectable and 21 were not. The tumor contours are supplied by a medical expert. We present an approach that uses intensity, shape, and texture features to determine tumor resectability. The best classification results are obtained with fine Gaussian SVM and the L0 Feature Selection algorithms. Compared to expert predictions made on the same dataset, our method achieves better classification results. We obtain significantly better results on correctly predicting non-resectability (+17%) compared to a expert, which is essential for patient treatment (negative prediction value). Moreover, our predictions of resectability exceed expert predictions by approximately 3% (positive prediction value).

Original languageEnglish
Title of host publicationSPIE.Medical Imaging: Computer-Aided Diagnosis, 10-15 February 2018, Houston, Texas
Subtitle of host publicationComputer-Aided Diagnosis
Place of PublicationBellingham
PublisherSPIE
Pages1-12
Volume10575
ISBN (Electronic)9781510616394
DOIs
Publication statusPublished - 1 Jan 2018
Event2018 SPIE Medical Imaging: Image Processing - Houston, United States
Duration: 10 Feb 201815 Feb 2018

Publication series

NameProceedings of SPIE
Volume1057530

Conference

Conference2018 SPIE Medical Imaging: Image Processing
CountryUnited States
CityHouston
Period10/02/1815/02/18

Fingerprint

Adenocarcinoma
Tumors
tumors
predictions
Neoplasms
Patient treatment
physicians
Feature extraction
Physicians
textures
Textures
Therapeutics

Bibliographical note

session PS9

Keywords

  • Classification
  • Computer-Aided Diagnosis
  • CT
  • Pancreatic ductal adenocarcinoma
  • Radiomics
  • Resectability Prediction

Cite this

van der Putten, J., Zinger, S., van der Sommen, F., de With, P. H. N., Prokop, M., & Hermans, J. (2018). Quantitative CT based radiomics as predictor of resectability of pancreatic adenocarcinoma. In SPIE.Medical Imaging: Computer-Aided Diagnosis, 10-15 February 2018, Houston, Texas : Computer-Aided Diagnosis (Vol. 10575, pp. 1-12). [105753O] (Proceedings of SPIE; Vol. 1057530). Bellingham: SPIE. https://doi.org/10.1117/12.2291746
van der Putten, J. ; Zinger, S. ; van der Sommen, F. ; de With, P.H.N. ; Prokop, M. ; Hermans, J. / Quantitative CT based radiomics as predictor of resectability of pancreatic adenocarcinoma. SPIE.Medical Imaging: Computer-Aided Diagnosis, 10-15 February 2018, Houston, Texas : Computer-Aided Diagnosis. Vol. 10575 Bellingham : SPIE, 2018. pp. 1-12 (Proceedings of SPIE).
@inproceedings{4e22b58cf6c84c0a8dc9452e53579dbf,
title = "Quantitative CT based radiomics as predictor of resectability of pancreatic adenocarcinoma",
abstract = "In current clinical practice, the resectability of pancreatic ductal adenocarcinoma (PDA) is determined subjec-tively by a physician, which is an error-prone procedure. In this paper, we present a method for automated determination of resectability of PDA from a routine abdominal CT, to reduce such decision errors. The tumor features are extracted from a group of patients with both hypo- A nd iso-attenuating tumors, of which 29 were resectable and 21 were not. The tumor contours are supplied by a medical expert. We present an approach that uses intensity, shape, and texture features to determine tumor resectability. The best classification results are obtained with fine Gaussian SVM and the L0 Feature Selection algorithms. Compared to expert predictions made on the same dataset, our method achieves better classification results. We obtain significantly better results on correctly predicting non-resectability (+17{\%}) compared to a expert, which is essential for patient treatment (negative prediction value). Moreover, our predictions of resectability exceed expert predictions by approximately 3{\%} (positive prediction value).",
keywords = "Classification, Computer-Aided Diagnosis, CT, Pancreatic ductal adenocarcinoma, Radiomics, Resectability Prediction",
author = "{van der Putten}, J. and S. Zinger and {van der Sommen}, F. and {de With}, P.H.N. and M. Prokop and J. Hermans",
note = "session PS9",
year = "2018",
month = "1",
day = "1",
doi = "10.1117/12.2291746",
language = "English",
volume = "10575",
series = "Proceedings of SPIE",
publisher = "SPIE",
pages = "1--12",
booktitle = "SPIE.Medical Imaging: Computer-Aided Diagnosis, 10-15 February 2018, Houston, Texas",
address = "United States",

}

van der Putten, J, Zinger, S, van der Sommen, F, de With, PHN, Prokop, M & Hermans, J 2018, Quantitative CT based radiomics as predictor of resectability of pancreatic adenocarcinoma. in SPIE.Medical Imaging: Computer-Aided Diagnosis, 10-15 February 2018, Houston, Texas : Computer-Aided Diagnosis. vol. 10575, 105753O, Proceedings of SPIE, vol. 1057530, SPIE, Bellingham, pp. 1-12, 2018 SPIE Medical Imaging: Image Processing, Houston, United States, 10/02/18. https://doi.org/10.1117/12.2291746

Quantitative CT based radiomics as predictor of resectability of pancreatic adenocarcinoma. / van der Putten, J.; Zinger, S.; van der Sommen, F.; de With, P.H.N.; Prokop, M.; Hermans, J.

SPIE.Medical Imaging: Computer-Aided Diagnosis, 10-15 February 2018, Houston, Texas : Computer-Aided Diagnosis. Vol. 10575 Bellingham : SPIE, 2018. p. 1-12 105753O (Proceedings of SPIE; Vol. 1057530).

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Quantitative CT based radiomics as predictor of resectability of pancreatic adenocarcinoma

AU - van der Putten, J.

AU - Zinger, S.

AU - van der Sommen, F.

AU - de With, P.H.N.

AU - Prokop, M.

AU - Hermans, J.

N1 - session PS9

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In current clinical practice, the resectability of pancreatic ductal adenocarcinoma (PDA) is determined subjec-tively by a physician, which is an error-prone procedure. In this paper, we present a method for automated determination of resectability of PDA from a routine abdominal CT, to reduce such decision errors. The tumor features are extracted from a group of patients with both hypo- A nd iso-attenuating tumors, of which 29 were resectable and 21 were not. The tumor contours are supplied by a medical expert. We present an approach that uses intensity, shape, and texture features to determine tumor resectability. The best classification results are obtained with fine Gaussian SVM and the L0 Feature Selection algorithms. Compared to expert predictions made on the same dataset, our method achieves better classification results. We obtain significantly better results on correctly predicting non-resectability (+17%) compared to a expert, which is essential for patient treatment (negative prediction value). Moreover, our predictions of resectability exceed expert predictions by approximately 3% (positive prediction value).

AB - In current clinical practice, the resectability of pancreatic ductal adenocarcinoma (PDA) is determined subjec-tively by a physician, which is an error-prone procedure. In this paper, we present a method for automated determination of resectability of PDA from a routine abdominal CT, to reduce such decision errors. The tumor features are extracted from a group of patients with both hypo- A nd iso-attenuating tumors, of which 29 were resectable and 21 were not. The tumor contours are supplied by a medical expert. We present an approach that uses intensity, shape, and texture features to determine tumor resectability. The best classification results are obtained with fine Gaussian SVM and the L0 Feature Selection algorithms. Compared to expert predictions made on the same dataset, our method achieves better classification results. We obtain significantly better results on correctly predicting non-resectability (+17%) compared to a expert, which is essential for patient treatment (negative prediction value). Moreover, our predictions of resectability exceed expert predictions by approximately 3% (positive prediction value).

KW - Classification

KW - Computer-Aided Diagnosis

KW - CT

KW - Pancreatic ductal adenocarcinoma

KW - Radiomics

KW - Resectability Prediction

UR - http://www.scopus.com/inward/record.url?scp=85046293531&partnerID=8YFLogxK

U2 - 10.1117/12.2291746

DO - 10.1117/12.2291746

M3 - Conference contribution

AN - SCOPUS:85046293531

VL - 10575

T3 - Proceedings of SPIE

SP - 1

EP - 12

BT - SPIE.Medical Imaging: Computer-Aided Diagnosis, 10-15 February 2018, Houston, Texas

PB - SPIE

CY - Bellingham

ER -

van der Putten J, Zinger S, van der Sommen F, de With PHN, Prokop M, Hermans J. Quantitative CT based radiomics as predictor of resectability of pancreatic adenocarcinoma. In SPIE.Medical Imaging: Computer-Aided Diagnosis, 10-15 February 2018, Houston, Texas : Computer-Aided Diagnosis. Vol. 10575. Bellingham: SPIE. 2018. p. 1-12. 105753O. (Proceedings of SPIE). https://doi.org/10.1117/12.2291746