EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Distinguishing pure histopathological growth patterns of colorectal liver metastases on CT using deep learning and radiomics: a pilot study

Martijn P A Starmans iD, Florian E Buisman, Michel Renckens, François E J A Willemssen, Sebastian R van der Voort, Bas Groot Koerkamp, Dirk J Grünhagen, Wiro J Niessen, Peter B Vermeulen, Cornelis Verhoef, Jacob J Visser, Stefan Klein

Clinical & Experimental Metastasis · 2021-10 · 38 (5) · 483-494

DOI: 10.1007/s10585-021-10119-6 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Clinical & Experimental Metastasis, online erschienen am 17.09.2021. Verfasst von Martijn P A Starmans und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology and Nuclear Medicine, Erasmus MC, Rotterdam, The Netherlands. m.starmans@erasmusmc.nl. Hauptthemen laut MEDLINE: Deep Learning. Die Arbeit wurde bisher 39-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Stichting voor de Technische Wetenschappen, KWF Kankerbestrijding.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Histopathological growth patterns (HGPs) are independent prognosticators for colorectal liver metastases (CRLM). Currently, HGPs are determined postoperatively. In this study, we evaluated radiomics for preoperative prediction of HGPs on computed tomography (CT), and its robustness to segmentation and acquisition variations. Patients with pure HGPs [i.e. 100% desmoplastic (dHGP) or 100% replacement (rHGP)] and a CT-scan who were surgically treated at the Erasmus MC between 2003–2015 were included retrospectively. Each lesion was segmented by three clinicians and a convolutional neural network (CNN). A prediction model was created using 564 radiomics features and a combination of machine learning approaches by training on the clinician’s and testing on the unseen CNN segmentations. The intra-class correlation coefficient (ICC) was used to select features robust to segmentation variations; ComBat was used to harmonize for acquisition variations. Evaluation was performed through a 100 × random-split cross-validation. The study included 93 CRLM in 76 patients (48% dHGP; 52% rHGP). Despite substantial differences between the segmentations of the three clinicians and the CNN, the radiomics model had a mean area under the curve of 0.69. ICC-based feature selection or ComBat yielded no improvement. Concluding, the combination of a CNN for segmentation and radiomics for classification has potential for automatically distinguishing dHGPs from rHGP, and is robust to segmentation and acquisition variations. Pending further optimization, including extension to mixed HGPs, our model may serve as a preoperative addition to postoperative HGP assessment, enabling further exploitation of HGPs as a biomarker.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningAgedColorectal NeoplasmsFemaleHumansLiver NeoplasmsMaleMiddle AgedPilot ProjectsTomography, X-Ray Computed

Stichwörter der Autor:innen

BiomarkersComputedDeep learningLiver neoplasmsMachine learningTomographyX-ray

Autor:innen und Institutionen

  • Department of Radiology and Nuclear Medicine, Erasmus MC, Rotterdam, The Netherlands. m.starmans@erasmusmc.nl.
  • Department of Surgery, Erasmus MC Cancer Institute, Rotterdam, The Netherlands.
  • Department of Radiology and Nuclear Medicine, Erasmus MC, Rotterdam, The Netherlands.
  • Faculty of Applied Sciences, Delft University of Technology, Delft, The Netherlands.
  • Translational Cancer Research Unit, Department of Oncological Research, Oncology Center, GZA Hospitals Campus Sint-Augustinus and University of Antwerp, Antwerp, Belgium.

ORCID-Profile: Martijn P A Starmans

Förderung

  • Stichting voor de Technische Wetenschappen
  • KWF Kankerbestrijding

Zitieren

APA

Starmans, M. P. A., Buisman, F. E., Renckens, M., Willemssen, F. E. J. A., van der Voort, S. R., Groot Koerkamp, B., Grünhagen, D. J., Niessen, W. J., Vermeulen, P. B., Verhoef, C., Visser, J. J., & Klein, S. (2021). Distinguishing pure histopathological growth patterns of colorectal liver metastases on CT using deep learning and radiomics: a pilot study. Clinical & Experimental Metastasis, 38(5), 483-494. https://doi.org/10.1007/s10585-021-10119-6

Vancouver

Starmans MPA, Buisman FE, Renckens M, Willemssen FEJA, van der Voort SR, Groot Koerkamp B, et al. Distinguishing pure histopathological growth patterns of colorectal liver metastases on CT using deep learning and radiomics: a pilot study. Clinical & Experimental Metastasis. 2021;38(5):483-494. doi:10.1007/s10585-021-10119-6

BibTeX
@article{Starmans2021,
  author = {Starmans, Martijn P A and Buisman, Florian E and Renckens, Michel and Willemssen, François E J A and van der Voort, Sebastian R and Groot Koerkamp, Bas and Grünhagen, Dirk J and Niessen, Wiro J and Vermeulen, Peter B and Verhoef, Cornelis and Visser, Jacob J and Klein, Stefan},
  title = {Distinguishing pure histopathological growth patterns of colorectal liver metastases on CT using deep learning and radiomics: a pilot study},
  journal = {Clinical & Experimental Metastasis},
  year = {2021},
  volume = {38},
  number = {5},
  pages = {483--494},
  doi = {10.1007/s10585-021-10119-6}
}
Kostenlos starten →