A radiomics approach for lung nodule detection in thoracic CT images based on the dynamic patterns of morphological variation

被引:4
作者
Lin, Fan-Ya [1 ]
Chang, Yeun-Chung [2 ,3 ]
Huang, Hsuan-Yu [5 ]
Li, Chia-Chen [1 ]
Chen, Yi-Chang [1 ,4 ]
Chen, Chung-Ming [1 ]
机构
[1] Natl Taiwan Univ, Coll Med & Coll Engn, Dept Biomed Engn, 1,Sec 1,Jen Ai Rd, Taipei 100, Taiwan
[2] Natl Taiwan Univ Hosp, Dept Med Imaging, Taipei, Taiwan
[3] Natl Taiwan Univ, Coll Med, Taipei, Taiwan
[4] Cardinal Tien Hosp, Dept Med Imaging, New Taipei, Taiwan
[5] IOIP Taiwan CO Ltd, Taipei, Taiwan
关键词
Lung; X-ray computed tomography; Multiple pulmonary nodules; Machine learning; Deep learning; FALSE-POSITIVE REDUCTION; COMPUTER-AIDED DETECTION; PULMONARY NODULES; AUTOMATIC DETECTION; DETECTION SYSTEM; VARIABILITY; ALGORITHMS; SCANS;
D O I
10.1007/s00330-021-08456-x
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Objectives To propose and evaluate a set of radiomic features, called morphological dynamics features, for pulmonary nodule detection, which were rooted in the dynamic patterns of morphological variation and needless precise lesion segmentation. Materials and methods Two datasets were involved, namely, university hospital (UH) and LIDC datasets, comprising 72 CT scans (360 nodules) and 888 CT scans (2230 nodules), respectively. Each nodule was annotated by multiple radiologists. Denoted the category of nodules identified by at least k radiologists as ALk. A nodule detection algorithm, called CAD-MD algorithm, was proposed based on the morphological dynamics radiomic features, characterizing a lesion by ten sets of the same features with different values extracted from ten different thresholding results. Each nodule candidate was classified by a two-level classifier, including ten decision trees and a random forest, respectively. The CAD-MD algorithm was compared with a deep learning approach, the N-Net, using the UH dataset. Results On the AL1 and AL2 of the UH dataset, the AUC of the AFROC curves were 0.777 and 0.851 for the CAD-MD algorithm and 0.478 and 0.472 for the N-Net, respectively. The CAD-MD algorithm achieved the sensitivities of 84.4% and 91.4% with 2.98 and 3.69 FPs/scan and the N-Net 74.4% and 80.7% with 3.90 and 4.49 FPs/scan, respectively. On the LIDC dataset, the CAD-MD algorithm attained the sensitivities of 87.6%, 89.2%, 92.2%, and 95.0% with 4 FPs/scan for AL1-AL4, respectively. Conclusion The morphological dynamics radiomic features might serve as an effective set of radiomic features for lung nodule detection.
引用
收藏
页码:3767 / 3777
页数:11
相关论文
共 60 条
[1]   Lung Nodule Detection via Deep Reinforcement Learning [J].
Ali, Issa ;
Hart, Gregory R. ;
Gunabushanam, Gowthaman ;
Liang, Ying ;
Muhammad, Wazir ;
Nartowt, Bradley ;
Kane, Michael ;
Ma, Xiaomei ;
Deng, Jun .
FRONTIERS IN ONCOLOGY, 2018, 8
[2]  
[Anonymous], 2011, LUNG IMAGE DATABASE
[3]   End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography [J].
Ardila, Diego ;
Kiraly, Atilla P. ;
Bharadwaj, Sujeeth ;
Choi, Bokyung ;
Reicher, Joshua J. ;
Peng, Lily ;
Tse, Daniel ;
Etemadi, Mozziyar ;
Ye, Wenxing ;
Corrado, Greg ;
Naidich, David P. ;
Shetty, Shravya .
NATURE MEDICINE, 2019, 25 (06) :954-+
[4]   The Lung Image Database Consortium, (LIDC) and Image Database Resource Initiative (IDRI): A Completed Reference Database of Lung Nodules on CT Scans [J].
Armato, Samuel G., III ;
McLennan, Geoffrey ;
Bidaut, Luc ;
McNitt-Gray, Michael F. ;
Meyer, Charles R. ;
Reeves, Anthony P. ;
Zhao, Binsheng ;
Aberle, Denise R. ;
Henschke, Claudia I. ;
Hoffman, Eric A. ;
Kazerooni, Ella A. ;
MacMahon, Heber ;
van Beek, Edwin J. R. ;
Yankelevitz, David ;
Biancardi, Alberto M. ;
Bland, Peyton H. ;
Brown, Matthew S. ;
Engelmann, Roger M. ;
Laderach, Gary E. ;
Max, Daniel ;
Pais, Richard C. ;
Qing, David P-Y ;
Roberts, Rachael Y. ;
Smith, Amanda R. ;
Starkey, Adam ;
Batra, Poonam ;
Caligiuri, Philip ;
Farooqi, Ali ;
Gladish, Gregory W. ;
Jude, C. Matilda ;
Munden, Reginald F. ;
Petkovska, Iva ;
Quint, Leslie E. ;
Schwartz, Lawrence H. ;
Sundaram, Baskaran ;
Dodd, Lori E. ;
Fenimore, Charles ;
Gur, David ;
Petrick, Nicholas ;
Freymann, John ;
Kirby, Justin ;
Hughes, Brian ;
Casteele, Alessi Vande ;
Gupte, Sangeeta ;
Sallam, Maha ;
Heath, Michael D. ;
Kuhn, Michael H. ;
Dharaiya, Ekta ;
Burns, Richard ;
Fryd, David S. .
MEDICAL PHYSICS, 2011, 38 (02) :915-931
[5]   Automated detection of lung nodules in CT scans: Preliminary results [J].
Armato, SG ;
Giger, ML ;
MacMahon, H .
MEDICAL PHYSICS, 2001, 28 (08) :1552-1561
[6]   Deep convolutional networks do not classify based on global object shape [J].
Baker, Nicholas ;
Lu, Hongjing ;
Erlikhman, Gennady ;
Kellman, Philip J. .
PLOS COMPUTATIONAL BIOLOGY, 2018, 14 (12)
[7]   Semi-automated pulmonary nodule interval segmentation using the NLST data [J].
Balagurunathan, Yoganand ;
Beers, Andrew ;
Kalpathy-Cramer, Jayashree ;
McNitt-Gray, Michael ;
Hadjiiski, Lubomir ;
Zhao, Bensheng ;
Zhu, Jiangguo ;
Yang, Hao ;
Yip, Stephen S. F. ;
Aerts, Hugo J. W. L. ;
Napel, Sandy ;
Cherezov, Dmitrii ;
Cha, Kenny ;
Chan, Heang-Ping ;
Flores, Carlos ;
Garcia, Alberto ;
Gillies, Robert ;
Goldgof, Dmitry .
MEDICAL PHYSICS, 2018, 45 (03) :1093-1107
[8]   Radiomics of CT Features May Be Nonreproducible and Redundant: Influence of CT Acquisition Parameters [J].
Berenguer, Roberto ;
del Rosario Pastor-Juan, Maria ;
Canales-Vazquez, Jesus ;
Castro-Garcia, Miguel ;
Villas, Maria Victoria ;
Mansilla Legorburo, Francisco ;
Sabater, Sebastia .
RADIOLOGY, 2018, 288 (02) :407-415
[9]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[10]  
Brochu F., 2019, ARXIV PREPRINT ARXIV