Perinodular and Intranodular Radiomic Features on Lung CT Images Distinguish Adenocarcinomas from Granulomas

Niha Beig, Mohammadhadi Khorrami, Mehdi Alilou, Prateek Prasanna, Nathaniel Braman, Mahdi Orooji, Sagar Rakshit, Kaustav Bera, Prabhakar Rajiah, Jennifer Ginsberg, Christopher Donatelli, Rajat Thawani, Michael Yang, Frank Jacono, Pallavi Tiwari, Vamsidhar Velcheti, Robert Gilkeson, Philip Linden, Anant Madabhushi

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Purpose To evaluate ability of radiomic (computer-extracted imaging) features to distinguish non-small cell lung cancer adenocarcinomas from granulomas at noncontrast CT. Materials and Methods For this retrospective study, screening or standard diagnostic noncontrast CT images were collected for 290 patients (mean age, 68 years; range, 18-92 years; 125 men [mean age, 67 years; range, 18-90 years] and 165 women [mean age, 68 years; range, 33-92 years]) from two institutions between 2007 and 2013. Histopathologic analysis was available for one nodule per patient. Corresponding nodule of interest was identified on axial CT images by a radiologist with manual annotation. Nodule shape, wavelet (Gabor), and texture-based (Haralick and Laws energy) features were extracted from intra- and perinodular regions. Features were pruned to train machine learning classifiers with 145 patients. In a test set of 145 patients, classifier results were compared against a convolutional neural network (CNN) and diagnostic readings of two radiologists. Results Support vector machine classifier with intranodular radiomic features achieved an area under the receiver operating characteristic curve (AUC) of 0.75 on the test set. Combining radiomics of intranodular with perinodular regions improved the AUC to 0.80. On the same test set, CNN resulted in an AUC of 0.76. Radiologist readers achieved AUCs of 0.61 and 0.60, respectively. Conclusion Radiomic features from intranodular and perinodular regions of nodules can distinguish non-small cell lung cancer adenocarcinomas from benign granulomas at noncontrast CT.

Original languageEnglish (US)
Pages (from-to)783-792
Number of pages10
JournalRadiology
Volume290
Issue number3
DOIs
StatePublished - Mar 1 2019

Fingerprint

Granuloma
Area Under Curve
Adenocarcinoma
Lung
Non-Small Cell Lung Carcinoma
ROC Curve
Reading
Retrospective Studies
Radiologists
Adenocarcinoma of lung

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Beig, N., Khorrami, M., Alilou, M., Prasanna, P., Braman, N., Orooji, M., ... Madabhushi, A. (2019). Perinodular and Intranodular Radiomic Features on Lung CT Images Distinguish Adenocarcinomas from Granulomas. Radiology, 290(3), 783-792. https://doi.org/10.1148/radiol.2018180910

Perinodular and Intranodular Radiomic Features on Lung CT Images Distinguish Adenocarcinomas from Granulomas. / Beig, Niha; Khorrami, Mohammadhadi; Alilou, Mehdi; Prasanna, Prateek; Braman, Nathaniel; Orooji, Mahdi; Rakshit, Sagar; Bera, Kaustav; Rajiah, Prabhakar; Ginsberg, Jennifer; Donatelli, Christopher; Thawani, Rajat; Yang, Michael; Jacono, Frank; Tiwari, Pallavi; Velcheti, Vamsidhar; Gilkeson, Robert; Linden, Philip; Madabhushi, Anant.

In: Radiology, Vol. 290, No. 3, 01.03.2019, p. 783-792.

Research output: Contribution to journalArticle

Beig, N, Khorrami, M, Alilou, M, Prasanna, P, Braman, N, Orooji, M, Rakshit, S, Bera, K, Rajiah, P, Ginsberg, J, Donatelli, C, Thawani, R, Yang, M, Jacono, F, Tiwari, P, Velcheti, V, Gilkeson, R, Linden, P & Madabhushi, A 2019, 'Perinodular and Intranodular Radiomic Features on Lung CT Images Distinguish Adenocarcinomas from Granulomas', Radiology, vol. 290, no. 3, pp. 783-792. https://doi.org/10.1148/radiol.2018180910
Beig, Niha ; Khorrami, Mohammadhadi ; Alilou, Mehdi ; Prasanna, Prateek ; Braman, Nathaniel ; Orooji, Mahdi ; Rakshit, Sagar ; Bera, Kaustav ; Rajiah, Prabhakar ; Ginsberg, Jennifer ; Donatelli, Christopher ; Thawani, Rajat ; Yang, Michael ; Jacono, Frank ; Tiwari, Pallavi ; Velcheti, Vamsidhar ; Gilkeson, Robert ; Linden, Philip ; Madabhushi, Anant. / Perinodular and Intranodular Radiomic Features on Lung CT Images Distinguish Adenocarcinomas from Granulomas. In: Radiology. 2019 ; Vol. 290, No. 3. pp. 783-792.
@article{f82e345ba20340d985b70101ff5ee1c5,
title = "Perinodular and Intranodular Radiomic Features on Lung CT Images Distinguish Adenocarcinomas from Granulomas",
abstract = "Purpose To evaluate ability of radiomic (computer-extracted imaging) features to distinguish non-small cell lung cancer adenocarcinomas from granulomas at noncontrast CT. Materials and Methods For this retrospective study, screening or standard diagnostic noncontrast CT images were collected for 290 patients (mean age, 68 years; range, 18-92 years; 125 men [mean age, 67 years; range, 18-90 years] and 165 women [mean age, 68 years; range, 33-92 years]) from two institutions between 2007 and 2013. Histopathologic analysis was available for one nodule per patient. Corresponding nodule of interest was identified on axial CT images by a radiologist with manual annotation. Nodule shape, wavelet (Gabor), and texture-based (Haralick and Laws energy) features were extracted from intra- and perinodular regions. Features were pruned to train machine learning classifiers with 145 patients. In a test set of 145 patients, classifier results were compared against a convolutional neural network (CNN) and diagnostic readings of two radiologists. Results Support vector machine classifier with intranodular radiomic features achieved an area under the receiver operating characteristic curve (AUC) of 0.75 on the test set. Combining radiomics of intranodular with perinodular regions improved the AUC to 0.80. On the same test set, CNN resulted in an AUC of 0.76. Radiologist readers achieved AUCs of 0.61 and 0.60, respectively. Conclusion Radiomic features from intranodular and perinodular regions of nodules can distinguish non-small cell lung cancer adenocarcinomas from benign granulomas at noncontrast CT.",
author = "Niha Beig and Mohammadhadi Khorrami and Mehdi Alilou and Prateek Prasanna and Nathaniel Braman and Mahdi Orooji and Sagar Rakshit and Kaustav Bera and Prabhakar Rajiah and Jennifer Ginsberg and Christopher Donatelli and Rajat Thawani and Michael Yang and Frank Jacono and Pallavi Tiwari and Vamsidhar Velcheti and Robert Gilkeson and Philip Linden and Anant Madabhushi",
year = "2019",
month = "3",
day = "1",
doi = "10.1148/radiol.2018180910",
language = "English (US)",
volume = "290",
pages = "783--792",
journal = "Radiology",
issn = "0033-8419",
publisher = "Radiological Society of North America Inc.",
number = "3",

}

TY - JOUR

T1 - Perinodular and Intranodular Radiomic Features on Lung CT Images Distinguish Adenocarcinomas from Granulomas

AU - Beig, Niha

AU - Khorrami, Mohammadhadi

AU - Alilou, Mehdi

AU - Prasanna, Prateek

AU - Braman, Nathaniel

AU - Orooji, Mahdi

AU - Rakshit, Sagar

AU - Bera, Kaustav

AU - Rajiah, Prabhakar

AU - Ginsberg, Jennifer

AU - Donatelli, Christopher

AU - Thawani, Rajat

AU - Yang, Michael

AU - Jacono, Frank

AU - Tiwari, Pallavi

AU - Velcheti, Vamsidhar

AU - Gilkeson, Robert

AU - Linden, Philip

AU - Madabhushi, Anant

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Purpose To evaluate ability of radiomic (computer-extracted imaging) features to distinguish non-small cell lung cancer adenocarcinomas from granulomas at noncontrast CT. Materials and Methods For this retrospective study, screening or standard diagnostic noncontrast CT images were collected for 290 patients (mean age, 68 years; range, 18-92 years; 125 men [mean age, 67 years; range, 18-90 years] and 165 women [mean age, 68 years; range, 33-92 years]) from two institutions between 2007 and 2013. Histopathologic analysis was available for one nodule per patient. Corresponding nodule of interest was identified on axial CT images by a radiologist with manual annotation. Nodule shape, wavelet (Gabor), and texture-based (Haralick and Laws energy) features were extracted from intra- and perinodular regions. Features were pruned to train machine learning classifiers with 145 patients. In a test set of 145 patients, classifier results were compared against a convolutional neural network (CNN) and diagnostic readings of two radiologists. Results Support vector machine classifier with intranodular radiomic features achieved an area under the receiver operating characteristic curve (AUC) of 0.75 on the test set. Combining radiomics of intranodular with perinodular regions improved the AUC to 0.80. On the same test set, CNN resulted in an AUC of 0.76. Radiologist readers achieved AUCs of 0.61 and 0.60, respectively. Conclusion Radiomic features from intranodular and perinodular regions of nodules can distinguish non-small cell lung cancer adenocarcinomas from benign granulomas at noncontrast CT.

AB - Purpose To evaluate ability of radiomic (computer-extracted imaging) features to distinguish non-small cell lung cancer adenocarcinomas from granulomas at noncontrast CT. Materials and Methods For this retrospective study, screening or standard diagnostic noncontrast CT images were collected for 290 patients (mean age, 68 years; range, 18-92 years; 125 men [mean age, 67 years; range, 18-90 years] and 165 women [mean age, 68 years; range, 33-92 years]) from two institutions between 2007 and 2013. Histopathologic analysis was available for one nodule per patient. Corresponding nodule of interest was identified on axial CT images by a radiologist with manual annotation. Nodule shape, wavelet (Gabor), and texture-based (Haralick and Laws energy) features were extracted from intra- and perinodular regions. Features were pruned to train machine learning classifiers with 145 patients. In a test set of 145 patients, classifier results were compared against a convolutional neural network (CNN) and diagnostic readings of two radiologists. Results Support vector machine classifier with intranodular radiomic features achieved an area under the receiver operating characteristic curve (AUC) of 0.75 on the test set. Combining radiomics of intranodular with perinodular regions improved the AUC to 0.80. On the same test set, CNN resulted in an AUC of 0.76. Radiologist readers achieved AUCs of 0.61 and 0.60, respectively. Conclusion Radiomic features from intranodular and perinodular regions of nodules can distinguish non-small cell lung cancer adenocarcinomas from benign granulomas at noncontrast CT.

UR - http://www.scopus.com/inward/record.url?scp=85061984139&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061984139&partnerID=8YFLogxK

U2 - 10.1148/radiol.2018180910

DO - 10.1148/radiol.2018180910

M3 - Article

C2 - 30561278

AN - SCOPUS:85061984139

VL - 290

SP - 783

EP - 792

JO - Radiology

JF - Radiology

SN - 0033-8419

IS - 3

ER -