Automatic segmentation of the solid core and enclosed vessels in subsolid pulmonary nodules

Jean-Paul Charbonnier, Kaman Chung, Ernst T Scholten, Eva M van Rikxoort, Colin Jacobs, Nicola Sverzellati, Mario Silva, Ugo Pastorino, Bram van Ginneken, Francesco Ciompi

Research output: Contribution to journalArticle

Abstract

Subsolid pulmonary nodules are commonly encountered in lung cancer screening and clinical routine. Compared to other nodule types, subsolid nodules are associated with a higher malignancy probability for which the size and mass of the nodule and solid core are important indicators. However, reliably measuring these characteristics on computed tomography (CT) can be hampered by the presence of vessels encompassed by the nodule, since vessels have similar CT attenuation as solid cores. This can affect treatment decisions and patient management. We present a method based on voxel classification to automatically identify vessels and solid cores in given subsolid nodules on CT. Three experts validated our method on 170 screen-detected subsolid nodules from the Multicentric Italian Lung Disease trial. The agreement between the proposed method and the observers was substantial for vessel detection and moderate for solid core detection, which was similar to the inter-observer agreement. We found a relatively high variability in the inter-observer agreement and low method-observer agreements for delineating the borders of vessels and solid cores, illustrating the difficulty of this task. However, 92.4% of the proposed vessel and 80.6% of the proposed solid core segmentations were labeled as usable in clinical practice by the majority of experts.

Original languageEnglish
Pages (from-to)646
JournalScientific Reports
Volume8
Issue number1
DOIs
Publication statusPublished - Jan 12 2018

Fingerprint

Lung
Tomography
Observer Variation
Early Detection of Cancer
Lung Diseases
Lung Neoplasms
Neoplasms
Therapeutics

Keywords

  • Algorithms
  • Early Detection of Cancer
  • Humans
  • Lung Neoplasms/diagnostic imaging
  • Multiple Pulmonary Nodules/diagnostic imaging
  • Observer Variation
  • Reproducibility of Results
  • Tomography, X-Ray Computed/methods

Cite this

Charbonnier, J-P., Chung, K., Scholten, E. T., van Rikxoort, E. M., Jacobs, C., Sverzellati, N., ... Ciompi, F. (2018). Automatic segmentation of the solid core and enclosed vessels in subsolid pulmonary nodules. Scientific Reports, 8(1), 646. https://doi.org/10.1038/s41598-017-19101-3

Automatic segmentation of the solid core and enclosed vessels in subsolid pulmonary nodules. / Charbonnier, Jean-Paul; Chung, Kaman; Scholten, Ernst T; van Rikxoort, Eva M; Jacobs, Colin; Sverzellati, Nicola; Silva, Mario; Pastorino, Ugo; van Ginneken, Bram; Ciompi, Francesco.

In: Scientific Reports, Vol. 8, No. 1, 12.01.2018, p. 646.

Research output: Contribution to journalArticle

Charbonnier, J-P, Chung, K, Scholten, ET, van Rikxoort, EM, Jacobs, C, Sverzellati, N, Silva, M, Pastorino, U, van Ginneken, B & Ciompi, F 2018, 'Automatic segmentation of the solid core and enclosed vessels in subsolid pulmonary nodules', Scientific Reports, vol. 8, no. 1, pp. 646. https://doi.org/10.1038/s41598-017-19101-3
Charbonnier J-P, Chung K, Scholten ET, van Rikxoort EM, Jacobs C, Sverzellati N et al. Automatic segmentation of the solid core and enclosed vessels in subsolid pulmonary nodules. Scientific Reports. 2018 Jan 12;8(1):646. https://doi.org/10.1038/s41598-017-19101-3
Charbonnier, Jean-Paul ; Chung, Kaman ; Scholten, Ernst T ; van Rikxoort, Eva M ; Jacobs, Colin ; Sverzellati, Nicola ; Silva, Mario ; Pastorino, Ugo ; van Ginneken, Bram ; Ciompi, Francesco. / Automatic segmentation of the solid core and enclosed vessels in subsolid pulmonary nodules. In: Scientific Reports. 2018 ; Vol. 8, No. 1. pp. 646.
@article{5d1864dc816b4a5a9bed81977ef43494,
title = "Automatic segmentation of the solid core and enclosed vessels in subsolid pulmonary nodules",
abstract = "Subsolid pulmonary nodules are commonly encountered in lung cancer screening and clinical routine. Compared to other nodule types, subsolid nodules are associated with a higher malignancy probability for which the size and mass of the nodule and solid core are important indicators. However, reliably measuring these characteristics on computed tomography (CT) can be hampered by the presence of vessels encompassed by the nodule, since vessels have similar CT attenuation as solid cores. This can affect treatment decisions and patient management. We present a method based on voxel classification to automatically identify vessels and solid cores in given subsolid nodules on CT. Three experts validated our method on 170 screen-detected subsolid nodules from the Multicentric Italian Lung Disease trial. The agreement between the proposed method and the observers was substantial for vessel detection and moderate for solid core detection, which was similar to the inter-observer agreement. We found a relatively high variability in the inter-observer agreement and low method-observer agreements for delineating the borders of vessels and solid cores, illustrating the difficulty of this task. However, 92.4{\%} of the proposed vessel and 80.6{\%} of the proposed solid core segmentations were labeled as usable in clinical practice by the majority of experts.",
keywords = "Algorithms, Early Detection of Cancer, Humans, Lung Neoplasms/diagnostic imaging, Multiple Pulmonary Nodules/diagnostic imaging, Observer Variation, Reproducibility of Results, Tomography, X-Ray Computed/methods",
author = "Jean-Paul Charbonnier and Kaman Chung and Scholten, {Ernst T} and {van Rikxoort}, {Eva M} and Colin Jacobs and Nicola Sverzellati and Mario Silva and Ugo Pastorino and {van Ginneken}, Bram and Francesco Ciompi",
year = "2018",
month = "1",
day = "12",
doi = "10.1038/s41598-017-19101-3",
language = "English",
volume = "8",
pages = "646",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

TY - JOUR

T1 - Automatic segmentation of the solid core and enclosed vessels in subsolid pulmonary nodules

AU - Charbonnier, Jean-Paul

AU - Chung, Kaman

AU - Scholten, Ernst T

AU - van Rikxoort, Eva M

AU - Jacobs, Colin

AU - Sverzellati, Nicola

AU - Silva, Mario

AU - Pastorino, Ugo

AU - van Ginneken, Bram

AU - Ciompi, Francesco

PY - 2018/1/12

Y1 - 2018/1/12

N2 - Subsolid pulmonary nodules are commonly encountered in lung cancer screening and clinical routine. Compared to other nodule types, subsolid nodules are associated with a higher malignancy probability for which the size and mass of the nodule and solid core are important indicators. However, reliably measuring these characteristics on computed tomography (CT) can be hampered by the presence of vessels encompassed by the nodule, since vessels have similar CT attenuation as solid cores. This can affect treatment decisions and patient management. We present a method based on voxel classification to automatically identify vessels and solid cores in given subsolid nodules on CT. Three experts validated our method on 170 screen-detected subsolid nodules from the Multicentric Italian Lung Disease trial. The agreement between the proposed method and the observers was substantial for vessel detection and moderate for solid core detection, which was similar to the inter-observer agreement. We found a relatively high variability in the inter-observer agreement and low method-observer agreements for delineating the borders of vessels and solid cores, illustrating the difficulty of this task. However, 92.4% of the proposed vessel and 80.6% of the proposed solid core segmentations were labeled as usable in clinical practice by the majority of experts.

AB - Subsolid pulmonary nodules are commonly encountered in lung cancer screening and clinical routine. Compared to other nodule types, subsolid nodules are associated with a higher malignancy probability for which the size and mass of the nodule and solid core are important indicators. However, reliably measuring these characteristics on computed tomography (CT) can be hampered by the presence of vessels encompassed by the nodule, since vessels have similar CT attenuation as solid cores. This can affect treatment decisions and patient management. We present a method based on voxel classification to automatically identify vessels and solid cores in given subsolid nodules on CT. Three experts validated our method on 170 screen-detected subsolid nodules from the Multicentric Italian Lung Disease trial. The agreement between the proposed method and the observers was substantial for vessel detection and moderate for solid core detection, which was similar to the inter-observer agreement. We found a relatively high variability in the inter-observer agreement and low method-observer agreements for delineating the borders of vessels and solid cores, illustrating the difficulty of this task. However, 92.4% of the proposed vessel and 80.6% of the proposed solid core segmentations were labeled as usable in clinical practice by the majority of experts.

KW - Algorithms

KW - Early Detection of Cancer

KW - Humans

KW - Lung Neoplasms/diagnostic imaging

KW - Multiple Pulmonary Nodules/diagnostic imaging

KW - Observer Variation

KW - Reproducibility of Results

KW - Tomography, X-Ray Computed/methods

U2 - 10.1038/s41598-017-19101-3

DO - 10.1038/s41598-017-19101-3

M3 - Article

VL - 8

SP - 646

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

ER -