An integrated segmentation and shape-based classification scheme for distinguishing adenocarcinomas from granulomas on lung CT

Mehdi Alilou, Niha Beig, Mahdi Orooji, Prabhakar Rajiah, Vamsidhar Velcheti, Sagar Rakshit, Niyoti Reddy, Michael Yang, Frank Jacono, Robert C. Gilkeson, Philip Linden, Anant Madabhushi

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Purpose: Distinguishing between benign granulmoas and adenocarcinomas is confounded by their similar visual appearance on routine CT scans. Unfortunately, owing to the inability to discriminate these lesions radigraphically, many patients with benign granulomas are subjected to unnecessary surgical wedge resections and biopsies for pathologic confirmation of cancer presence or absence. This suggests the need for improved computerized characterization of these nodules in order to distinguish between these two classes of lesions on CT scans. While there has been substantial interest in the use of textural analysis for radiomic characterization of lung nodules, relatively less work has been done in shape based characterization of lung nodules, particularly with respect to granulmoas and adenocarcinomas. The primary goal of this study is to evaluate the role of 3D shape features for discrimination of benign granulomas from malignant adenocarcinomas on lung CT images. Towards this end we present an integrated framework for segmentation, feature characterization and classification of these nodules on CT. Methods: The nodule segmentation method starts with separation of lung regions from the surrounding lung anatomy. Next, the lung CT scans are projected into and represented in a three dimensional spectral embedding (SE) space, allowing for better determination of the boundaries of the nodule. This then enables the application of a gradient vector flow active contour (SEGvAC) model for nodule boundary extraction. A set of 24 shape features from both 2D slices and 3D surface of the segmented nodules are extracted, including features pertaining to the angularity, spiculation, elongation and nodule compactness. A feature selection scheme, PCA-VIP, is employed to identify the most discriminating set of features to distinguish granulmoas from adenocarcinomas within a learning set of 82 patients. The features thus identified were then combined with a support vector machine classifier and independently validated on a distinct test set comprising 67 patients. The performance of the classifier for both of the training and validation cohorts was evaluated by the area under receiver characteristic curve (ROC). Results: We used 82 and 67 studies from two different institutions respectively for training and independent validation of the model and the shape features. The Dice coefficient between automatically segmented nodules by SEGvAC and the manual delineations by expert radiologists (readers) was 0.84± 0.04 whereas inter-reader segmentation agreement was 0.79± 0.12. We also identified a set of consistent features (Roughness, Convexity and Spherecity) that were found to be strongly correlated across both manual and automated nodule segmentations (R > 0.80, p < 0.0001) and capture the marginal smoothness and 3D compactness of the nodules. On the independent validation set of 67 studies our classifier yielded a ROC AUC of 0.72 and 0.64 for manually- and automatically segmented nodules respectively. On a subset of 20 studies, the AUCs for the two expert radiologists and 1 pulmonologist were found to be 0.82, 0.68 and 0.58 respectively. Conclusions: The major finding of this study was that certain shape features appear to differentially express between granulomas and adenocarcinomas and thus computer extracted shape cues could be used to distinguish these radiographically similar pathologies.

Original languageEnglish (US)
Pages (from-to)3556-3569
Number of pages14
JournalMedical Physics
Volume44
Issue number7
DOIs
StatePublished - Jul 1 2017

Fingerprint

Granuloma
Adenocarcinoma
Lung
Area Under Curve
Passive Cutaneous Anaphylaxis
Hodgkin Disease
ROC Curve
Cues
Anatomy
Learning
Pathology
Biopsy
Neoplasms
Radiologists

Keywords

  • CADx
  • Lung CT
  • nodule characterization
  • segmentation
  • shape analysis

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

Alilou, M., Beig, N., Orooji, M., Rajiah, P., Velcheti, V., Rakshit, S., ... Madabhushi, A. (2017). An integrated segmentation and shape-based classification scheme for distinguishing adenocarcinomas from granulomas on lung CT. Medical Physics, 44(7), 3556-3569. https://doi.org/10.1002/mp.12208

An integrated segmentation and shape-based classification scheme for distinguishing adenocarcinomas from granulomas on lung CT. / Alilou, Mehdi; Beig, Niha; Orooji, Mahdi; Rajiah, Prabhakar; Velcheti, Vamsidhar; Rakshit, Sagar; Reddy, Niyoti; Yang, Michael; Jacono, Frank; Gilkeson, Robert C.; Linden, Philip; Madabhushi, Anant.

In: Medical Physics, Vol. 44, No. 7, 01.07.2017, p. 3556-3569.

Research output: Contribution to journalArticle

Alilou, M, Beig, N, Orooji, M, Rajiah, P, Velcheti, V, Rakshit, S, Reddy, N, Yang, M, Jacono, F, Gilkeson, RC, Linden, P & Madabhushi, A 2017, 'An integrated segmentation and shape-based classification scheme for distinguishing adenocarcinomas from granulomas on lung CT', Medical Physics, vol. 44, no. 7, pp. 3556-3569. https://doi.org/10.1002/mp.12208
Alilou, Mehdi ; Beig, Niha ; Orooji, Mahdi ; Rajiah, Prabhakar ; Velcheti, Vamsidhar ; Rakshit, Sagar ; Reddy, Niyoti ; Yang, Michael ; Jacono, Frank ; Gilkeson, Robert C. ; Linden, Philip ; Madabhushi, Anant. / An integrated segmentation and shape-based classification scheme for distinguishing adenocarcinomas from granulomas on lung CT. In: Medical Physics. 2017 ; Vol. 44, No. 7. pp. 3556-3569.
@article{7f5ea2c99c5348c79bd0725fded57ea8,
title = "An integrated segmentation and shape-based classification scheme for distinguishing adenocarcinomas from granulomas on lung CT",
abstract = "Purpose: Distinguishing between benign granulmoas and adenocarcinomas is confounded by their similar visual appearance on routine CT scans. Unfortunately, owing to the inability to discriminate these lesions radigraphically, many patients with benign granulomas are subjected to unnecessary surgical wedge resections and biopsies for pathologic confirmation of cancer presence or absence. This suggests the need for improved computerized characterization of these nodules in order to distinguish between these two classes of lesions on CT scans. While there has been substantial interest in the use of textural analysis for radiomic characterization of lung nodules, relatively less work has been done in shape based characterization of lung nodules, particularly with respect to granulmoas and adenocarcinomas. The primary goal of this study is to evaluate the role of 3D shape features for discrimination of benign granulomas from malignant adenocarcinomas on lung CT images. Towards this end we present an integrated framework for segmentation, feature characterization and classification of these nodules on CT. Methods: The nodule segmentation method starts with separation of lung regions from the surrounding lung anatomy. Next, the lung CT scans are projected into and represented in a three dimensional spectral embedding (SE) space, allowing for better determination of the boundaries of the nodule. This then enables the application of a gradient vector flow active contour (SEGvAC) model for nodule boundary extraction. A set of 24 shape features from both 2D slices and 3D surface of the segmented nodules are extracted, including features pertaining to the angularity, spiculation, elongation and nodule compactness. A feature selection scheme, PCA-VIP, is employed to identify the most discriminating set of features to distinguish granulmoas from adenocarcinomas within a learning set of 82 patients. The features thus identified were then combined with a support vector machine classifier and independently validated on a distinct test set comprising 67 patients. The performance of the classifier for both of the training and validation cohorts was evaluated by the area under receiver characteristic curve (ROC). Results: We used 82 and 67 studies from two different institutions respectively for training and independent validation of the model and the shape features. The Dice coefficient between automatically segmented nodules by SEGvAC and the manual delineations by expert radiologists (readers) was 0.84± 0.04 whereas inter-reader segmentation agreement was 0.79± 0.12. We also identified a set of consistent features (Roughness, Convexity and Spherecity) that were found to be strongly correlated across both manual and automated nodule segmentations (R > 0.80, p < 0.0001) and capture the marginal smoothness and 3D compactness of the nodules. On the independent validation set of 67 studies our classifier yielded a ROC AUC of 0.72 and 0.64 for manually- and automatically segmented nodules respectively. On a subset of 20 studies, the AUCs for the two expert radiologists and 1 pulmonologist were found to be 0.82, 0.68 and 0.58 respectively. Conclusions: The major finding of this study was that certain shape features appear to differentially express between granulomas and adenocarcinomas and thus computer extracted shape cues could be used to distinguish these radiographically similar pathologies.",
keywords = "CADx, Lung CT, nodule characterization, segmentation, shape analysis",
author = "Mehdi Alilou and Niha Beig and Mahdi Orooji and Prabhakar Rajiah and Vamsidhar Velcheti and Sagar Rakshit and Niyoti Reddy and Michael Yang and Frank Jacono and Gilkeson, {Robert C.} and Philip Linden and Anant Madabhushi",
year = "2017",
month = "7",
day = "1",
doi = "10.1002/mp.12208",
language = "English (US)",
volume = "44",
pages = "3556--3569",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "7",

}

TY - JOUR

T1 - An integrated segmentation and shape-based classification scheme for distinguishing adenocarcinomas from granulomas on lung CT

AU - Alilou, Mehdi

AU - Beig, Niha

AU - Orooji, Mahdi

AU - Rajiah, Prabhakar

AU - Velcheti, Vamsidhar

AU - Rakshit, Sagar

AU - Reddy, Niyoti

AU - Yang, Michael

AU - Jacono, Frank

AU - Gilkeson, Robert C.

AU - Linden, Philip

AU - Madabhushi, Anant

PY - 2017/7/1

Y1 - 2017/7/1

N2 - Purpose: Distinguishing between benign granulmoas and adenocarcinomas is confounded by their similar visual appearance on routine CT scans. Unfortunately, owing to the inability to discriminate these lesions radigraphically, many patients with benign granulomas are subjected to unnecessary surgical wedge resections and biopsies for pathologic confirmation of cancer presence or absence. This suggests the need for improved computerized characterization of these nodules in order to distinguish between these two classes of lesions on CT scans. While there has been substantial interest in the use of textural analysis for radiomic characterization of lung nodules, relatively less work has been done in shape based characterization of lung nodules, particularly with respect to granulmoas and adenocarcinomas. The primary goal of this study is to evaluate the role of 3D shape features for discrimination of benign granulomas from malignant adenocarcinomas on lung CT images. Towards this end we present an integrated framework for segmentation, feature characterization and classification of these nodules on CT. Methods: The nodule segmentation method starts with separation of lung regions from the surrounding lung anatomy. Next, the lung CT scans are projected into and represented in a three dimensional spectral embedding (SE) space, allowing for better determination of the boundaries of the nodule. This then enables the application of a gradient vector flow active contour (SEGvAC) model for nodule boundary extraction. A set of 24 shape features from both 2D slices and 3D surface of the segmented nodules are extracted, including features pertaining to the angularity, spiculation, elongation and nodule compactness. A feature selection scheme, PCA-VIP, is employed to identify the most discriminating set of features to distinguish granulmoas from adenocarcinomas within a learning set of 82 patients. The features thus identified were then combined with a support vector machine classifier and independently validated on a distinct test set comprising 67 patients. The performance of the classifier for both of the training and validation cohorts was evaluated by the area under receiver characteristic curve (ROC). Results: We used 82 and 67 studies from two different institutions respectively for training and independent validation of the model and the shape features. The Dice coefficient between automatically segmented nodules by SEGvAC and the manual delineations by expert radiologists (readers) was 0.84± 0.04 whereas inter-reader segmentation agreement was 0.79± 0.12. We also identified a set of consistent features (Roughness, Convexity and Spherecity) that were found to be strongly correlated across both manual and automated nodule segmentations (R > 0.80, p < 0.0001) and capture the marginal smoothness and 3D compactness of the nodules. On the independent validation set of 67 studies our classifier yielded a ROC AUC of 0.72 and 0.64 for manually- and automatically segmented nodules respectively. On a subset of 20 studies, the AUCs for the two expert radiologists and 1 pulmonologist were found to be 0.82, 0.68 and 0.58 respectively. Conclusions: The major finding of this study was that certain shape features appear to differentially express between granulomas and adenocarcinomas and thus computer extracted shape cues could be used to distinguish these radiographically similar pathologies.

AB - Purpose: Distinguishing between benign granulmoas and adenocarcinomas is confounded by their similar visual appearance on routine CT scans. Unfortunately, owing to the inability to discriminate these lesions radigraphically, many patients with benign granulomas are subjected to unnecessary surgical wedge resections and biopsies for pathologic confirmation of cancer presence or absence. This suggests the need for improved computerized characterization of these nodules in order to distinguish between these two classes of lesions on CT scans. While there has been substantial interest in the use of textural analysis for radiomic characterization of lung nodules, relatively less work has been done in shape based characterization of lung nodules, particularly with respect to granulmoas and adenocarcinomas. The primary goal of this study is to evaluate the role of 3D shape features for discrimination of benign granulomas from malignant adenocarcinomas on lung CT images. Towards this end we present an integrated framework for segmentation, feature characterization and classification of these nodules on CT. Methods: The nodule segmentation method starts with separation of lung regions from the surrounding lung anatomy. Next, the lung CT scans are projected into and represented in a three dimensional spectral embedding (SE) space, allowing for better determination of the boundaries of the nodule. This then enables the application of a gradient vector flow active contour (SEGvAC) model for nodule boundary extraction. A set of 24 shape features from both 2D slices and 3D surface of the segmented nodules are extracted, including features pertaining to the angularity, spiculation, elongation and nodule compactness. A feature selection scheme, PCA-VIP, is employed to identify the most discriminating set of features to distinguish granulmoas from adenocarcinomas within a learning set of 82 patients. The features thus identified were then combined with a support vector machine classifier and independently validated on a distinct test set comprising 67 patients. The performance of the classifier for both of the training and validation cohorts was evaluated by the area under receiver characteristic curve (ROC). Results: We used 82 and 67 studies from two different institutions respectively for training and independent validation of the model and the shape features. The Dice coefficient between automatically segmented nodules by SEGvAC and the manual delineations by expert radiologists (readers) was 0.84± 0.04 whereas inter-reader segmentation agreement was 0.79± 0.12. We also identified a set of consistent features (Roughness, Convexity and Spherecity) that were found to be strongly correlated across both manual and automated nodule segmentations (R > 0.80, p < 0.0001) and capture the marginal smoothness and 3D compactness of the nodules. On the independent validation set of 67 studies our classifier yielded a ROC AUC of 0.72 and 0.64 for manually- and automatically segmented nodules respectively. On a subset of 20 studies, the AUCs for the two expert radiologists and 1 pulmonologist were found to be 0.82, 0.68 and 0.58 respectively. Conclusions: The major finding of this study was that certain shape features appear to differentially express between granulomas and adenocarcinomas and thus computer extracted shape cues could be used to distinguish these radiographically similar pathologies.

KW - CADx

KW - Lung CT

KW - nodule characterization

KW - segmentation

KW - shape analysis

UR - http://www.scopus.com/inward/record.url?scp=85022064008&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85022064008&partnerID=8YFLogxK

U2 - 10.1002/mp.12208

DO - 10.1002/mp.12208

M3 - Article

C2 - 28295386

AN - SCOPUS:85022064008

VL - 44

SP - 3556

EP - 3569

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 7

ER -