Ming Li1, Vivek Narayan2, Ritu R Gill3, Jyothi P Jagannathan4, Maria F Barile3, Feng Gao1, Raphael Bueno5, Jagadeesan Jayender3. 1. 1 Department of Radiology, HuaDong Hospital, Fudan University, Shanghai, China. 2. 2 Dana Farber Cancer Institute, Boston, MA. 3. 3 Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, 75 Francis St, L1-050, Boston, MA 02115. 4. 4 Department of Radiology, Dana Farber Cancer Institute, Harvard Medical School, Boston, MA. 5. 5 Department of Surgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA.
Abstract
OBJECTIVE: The purposes of this study are to develop quantitative imaging biomarkers obtained from high-resolution CTs for classifying ground-glass nodules (GGNs) into atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC); to evaluate the utility of contrast enhancement for differential diagnosis; and to develop and validate a support vector machine (SVM) to predict the GGN type. MATERIALS AND METHODS: The heterogeneity of 248 GGNs was quantified using custom software. Statistical analysis with a univariate Kruskal-Wallis test was performed to evaluate metrics for significant differences among the four GGN groups. The heterogeneity metrics were used to train a SVM to learn and predict the lesion type. RESULTS: Fifty of 57 and 51 of 57 heterogeneity metrics showed statistically significant differences among the four GGN groups on unenhanced and contrast-enhanced CT scans, respectively. The SVM predicted lesion type with greater accuracy than did three expert radiologists. The accuracy of classifying the GGNs into the four groups on the basis of the SVM algorithm was 70.9%, whereas the accuracy of the radiologists was 39.6%. The accuracy of SVM in classifying the AIS and MIA nodules was 73.1%, and the accuracy of the radiologists was 35.7%. For indolent versus invasive lesions, the accuracy of the SVM was 88.1%, and the accuracy of the radiologists was 60.8%. We found that contrast enhancement does not significantly improve the differential diagnosis of GGNs. CONCLUSION: Compared with the GGN classification done by the three radiologists, the SVM trained regarding all the heterogeneity metrics showed significantly higher accuracy in classifying the lesions into the four groups, differentiating between AIS and MIA and between indolent and invasive lesions. Contrast enhancement did not improve the differential diagnosis of GGNs.
OBJECTIVE: The purposes of this study are to develop quantitative imaging biomarkers obtained from high-resolution CTs for classifying ground-glass nodules (GGNs) into atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC); to evaluate the utility of contrast enhancement for differential diagnosis; and to develop and validate a support vector machine (SVM) to predict the GGN type. MATERIALS AND METHODS: The heterogeneity of 248 GGNs was quantified using custom software. Statistical analysis with a univariate Kruskal-Wallis test was performed to evaluate metrics for significant differences among the four GGN groups. The heterogeneity metrics were used to train a SVM to learn and predict the lesion type. RESULTS: Fifty of 57 and 51 of 57 heterogeneity metrics showed statistically significant differences among the four GGN groups on unenhanced and contrast-enhanced CT scans, respectively. The SVM predicted lesion type with greater accuracy than did three expert radiologists. The accuracy of classifying the GGNs into the four groups on the basis of the SVM algorithm was 70.9%, whereas the accuracy of the radiologists was 39.6%. The accuracy of SVM in classifying the AIS and MIA nodules was 73.1%, and the accuracy of the radiologists was 35.7%. For indolent versus invasive lesions, the accuracy of the SVM was 88.1%, and the accuracy of the radiologists was 60.8%. We found that contrast enhancement does not significantly improve the differential diagnosis of GGNs. CONCLUSION: Compared with the GGN classification done by the three radiologists, the SVM trained regarding all the heterogeneity metrics showed significantly higher accuracy in classifying the lesions into the four groups, differentiating between AIS and MIA and between indolent and invasive lesions. Contrast enhancement did not improve the differential diagnosis of GGNs.
Entities:
Keywords:
International Association for the Study of Lung Cancer guidelines; computer-aided diagnosis; ground-glass opacity lesions; imaging biomarkers
Authors: Vishal K Patel; Sagar K Naik; David P Naidich; William D Travis; Jeremy A Weingarten; Richard Lazzaro; David D Gutterman; Catherine Wentowski; Horiana B Grosu; Suhail Raoof Journal: Chest Date: 2013-03 Impact factor: 9.410
Authors: Fabien Maldonado; Jennifer M Boland; Sushravya Raghunath; Marie Christine Aubry; Brian J Bartholmai; Mariza Deandrade; Thomas E Hartman; Ronald A Karwoski; Srinivasan Rajagopalan; Anne-Marie Sykes; Ping Yang; Eunhee S Yi; Richard A Robb; Tobias Peikert Journal: J Thorac Oncol Date: 2013-04 Impact factor: 15.609
Authors: Kyung Nyeo Jeon; Jin Mo Goo; Chang Hyun Lee; Youkyung Lee; Ji Yung Choo; Nyoung Keun Lee; Mi-Suk Shim; In Sun Lee; Kwang Gi Kim; David S Gierada; Kyongtae T Bae Journal: Invest Radiol Date: 2012-08 Impact factor: 6.016
Authors: David P Naidich; Alexander A Bankier; Heber MacMahon; Cornelia M Schaefer-Prokop; Massimo Pistolesi; Jin Mo Goo; Paolo Macchiarini; James D Crapo; Christian J Herold; John H Austin; William D Travis Journal: Radiology Date: 2012-10-15 Impact factor: 11.105
Authors: Alain C Borczuk; Fang Qian; Angeliki Kazeros; Jennifer Eleazar; Adel Assaad; Joshua R Sonett; Mark Ginsburg; Lyall Gorenstein; Charles A Powell Journal: Am J Surg Pathol Date: 2009-03 Impact factor: 6.394
Authors: Anastasia Oikonomou; Pascal Salazar; Yuchen Zhang; David M Hwang; Alexander Petersen; Adam A Dmytriw; Narinder S Paul; Elsie T Nguyen Journal: Sci Rep Date: 2019-04-12 Impact factor: 4.379
Authors: Dalia Fahmy; Heba Kandil; Adel Khelifi; Maha Yaghi; Mohammed Ghazal; Ahmed Sharafeldeen; Ali Mahmoud; Ayman El-Baz Journal: Cancers (Basel) Date: 2022-04-06 Impact factor: 6.639