Abstract
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.
Key Points
• Texture features varied with such CT system settings as reconstruction kernels of CT images, CT scanner models, and parameter settings, and so on.
• Shape and first-order statistics were shown to be the most robust features against variation in CT imaging parameters.
• The morphological dynamics radiomic features, which mainly characterized the dynamic patterns of morphological variation, were shown to be effective for lung nodule detection.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00330-021-08456-x/MediaObjects/330_2021_8456_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00330-021-08456-x/MediaObjects/330_2021_8456_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00330-021-08456-x/MediaObjects/330_2021_8456_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00330-021-08456-x/MediaObjects/330_2021_8456_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00330-021-08456-x/MediaObjects/330_2021_8456_Fig5_HTML.png)
Similar content being viewed by others
Abbreviations
- AFROC:
-
Alternative free-response receiver operating characteristic curves
- AFROC-AUC:
-
AUC of an AFROC curve
- AL:
-
Agreement level
- AUC:
-
Area under the curve
- BN:
-
Branch number
- CADe :
-
Computer-aided detection
- CAD-MD :
-
The computer-aided detection algorithm based on morphological dynamics
- CI:
-
Confidence interval
- CT :
-
Computed tomography
- ESM:
-
Erosion size map
- FP :
-
False positive
- HU:
-
Hounsfield unit
- IDRI:
-
Image Database Resource Initiative
- LIDC :
-
Lung Image Database Consortium
- LUNA16:
-
Lung Nodule Analysis 2016
- MHU:
-
Mean of Hounsfield unit
- MS:
-
Morphological size
- NC-ESM:
-
Erosion size map of nodule candidate
- NC-VOI:
-
Volume of interest of nodule candidate
- PFP:
-
The probability of detecting a false-positive in an image
- SC:
-
Shape coefficient
- SD:
-
Size difference
- UH :
-
University hospital
- VOI:
-
Volume of interest
References
Zhang G, Jiang S, Yang Z et al (2018) Automatic nodule detection for lung cancer in CT images: a review. Comput Biol Med 103:287–300
Pehrson LM, Nielsen MB, Ammitzbøl Lauridsen C (2019) Automatic pulmonary nodule detection applying deep learning or machine learning algorithms to the LIDC-IDRI database: a systematic review. Diagnostics (Basel) 9(1). doi: https://doi.org/10.3390/diagnostics9010029
Valente IR, Cortez PC, Neto EC et al (2016) Automatic 3D pulmonary nodule detection in CT images: a survey. Comput Methods Programs Biomed 124:91–107
Jacobs C, van Rikxoort EM, Twellmann T et al (2014) Automatic detection of subsolid pulmonary nodules in thoracic computed tomography images. Med Image Anal 18:374–384
Shaukat F, Raja G, Gooya A et al (2017) Fully automatic detection of lung nodules in CT images using a hybrid feature set. Med Phys 44(7):3615–3629
Naqi SM, Sharif M, Yasmin M (2018) Multistage segmentation model and SVM-ensemble for precise lung nodule detection. Int J Comput Assist Radiol Surg 13:1083–1095
Shaukat F, Raja G, Ashraf R, Khalid S, Ahmad M, Ali A (2019) Artificial neural network based classification of lung nodules in CT images using intensity, shape and texture features. J Ambient Intell Humaniz Comput 10:4135–4149
Naqi SM, Sharif M, Lali IU (2019) A 3D nodule candidate detection method supported by hybrid features to reduce false positives in lung nodule detection. Multimed Tools Appl 78:26287–26311
Khan SA, Nazir M, Khan MA et al (2019) Lungs nodule detection framework from computed tomography images using support vector machine. Microsc Res Tech 82:1256–1266
Khan SA, Hussain S, Yang S et al (2019) Effective and reliable framework for lung nodules detection from CT scan images. Sci Rep 9:4989
Gu Y, Lu X, Zhang B et al (2019) Automatic lung nodule detection using multiscale dot nodule-enhancement filter and weighted support vector machines in chest computed tomography. PLoS One 14(1):e0210551
Saba T (2019) Automated lung nodule detection and classification based on multiple classifiers voting. Microsc Res Tech 82:1601–1609
Huidrom R, Chanu YJ, Singh KM (2019) Pulmonary nodule detection on computed tomography using neuro-evolutionary scheme. Signal Image Video Process 13:53–60
He L, Huang Y, Ma Z et al (2016) Effects of contrast-enhancement, reconstruction slice thickness and convolution kernel on the diagnostic performance of radiomics signature in solitary pulmonary nodule. Sci Rep 6:34921
Setio AA, Ciompi F, Litjens G et al (2016) Pulmonary nodule detection in CT images: false positive reduction using multi-view convolutional networks. IEEE Trans Med Imaging 35(5):1160–1169
Dou Q, Chen H, Yu L et al (2017) Multilevel contextual 3-d CNNs for false positive reduction in pulmonary nodule detection. IEEE Trans Biomed Eng 64(7):1558–1567
Zhang J, Xia Y, Zeng H et al (2018) NODULe: Combining constrained multi-scale LoG filters with densely dilated 3D deep convolutional neural network for pulmonary nodule detection. Neurocomputing 317:159–167
Ali I, Gregory R. Hart GR, Gunabushanam G, et al (2018) Lung nodule detection via deep reinforcement learning. Front Oncol 8:Article 108
Nasrullah N, Sang J, Alam MS et al (2019) Automated lung nodule detection and classification using deep learning combined with multiple strategies. Sensors (Basel) 19(17):3722. https://doi.org/10.3390/s19173722
Xie H, Yang D, Sun N et al (2019) Automated pulmonary nodule detection in CT images using deep convolutional neural networks. Pattern Recognit 85:109–119
Ardila D, Kiraly AP, Bharadwaj S et al (2019) End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nat Med 25(6):954–961
Liao F, Liang M, Li Z et al (2019) Evaluate the malignancy of pulmonary nodules using the 3-D deep leaky noisy-OR network. IEEE Trans Neural Netw Learn Syst 30(11):3484–3495
Winkels M, Cohen TS (2019) Pulmonary nodule detection in CT scans with equivariant CNNs. Med Image Anal 55:15–26
Wang J, Wang J, Wen Y et al (2019) Pulmonary nodule detection in volumetric chest CT scans using CNNs-based nodule-size-adaptive detection and classification. IEEE Access 7:46033–46044
Zheng S, Guo J, Cui X et al (2020) Automatic pulmonary nodule detection in CT scans using convolutional neural networks based on maximum intensity projection. IEEE Trans Med Imaging 39(3):797–805
Kim H, Park CM, Lee M et al (2016) Impact of reconstruction algorithms on CT radiomic features of pulmonary tumors: analysis of intra- and inter-reader variability and inter-reconstruction algorithm variability. PLoS One 11:e0164924
Lo P, Young S, Kim HJ et al (2016) Variability in CT lung-nodule quantification: effects of dose reduction and reconstruction methods on density and texture based features. Med Phys 43:4854
Shafiq-ul-Hassan M, Zhang GG, Hunt DC et al (2018) Accounting for reconstruction kernel-induced variability in CT radiomic features using noise power spectra. J Med Imaging (Bellingham) 5:011013
Mackin D, Fave X, Zhang L et al (2015) Measuring computed tomography scanner variability of radiomics features. Invest Radiol 50:757–765
Zhao W, Zhang W, Sun Y et al (2019) Convolution kernel and iterative reconstruction affect the diagnostic performance of radiomics and deep learning in lung adenocarcinoma pathological subtypes. Thorac Cancer 10(10):1893–1903
Geirhos R, Rubisch P, Michaelis C, et al (2018) ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness. arXiv preprint arXiv:1811.12231
Baker N, Lu H, Erlikhman G et al (2018) Deep convolutional networks do not classify based on global object shape. PLoS Comput Biol 14(12):e1006613
Hermann KL, Kornblith S (2019) Exploring the origins and prevalence of texture bias in convolutional neural networks. arXiv preprint arXiv:1911.09071
Mackin D, Ger R, Gay S et al (2019) Matching and homogenizing convolution kernels for quantitative studies in computed tomography. Invest Radiol 54:288–295
Yang Y, Feng X, Chi W et al (2018) Deep learning aided decision support for pulmonary nodules diagnosing: a review. J Thorac Dis 10(Suppl 7):S867–S875
Kim M, Yun J, Cho Y et al (2019) Deep learning in medical imaging. Neurospine 16(4):657–668
Reiazi R, Abbas E, Famiyeh P et al (2021) The impact of the variation of imaging parameters on the robustness of computed tomography radiomic features: a review. Comput Biol Med 133:104400. https://doi.org/10.1016/j.compbiomed.2021.104400
Zhao B, Tan Y, Tsai WY et al (2016) Reproducibility of radiomics for deciphering tumor phenotype with imaging. Sci Rep 6:23428
Berenguer R, Pastor-Juan MDR, Canales-Vázquez J et al (2018) Radiomics of CT features may be nonreproducible and redundant: influence of CT acquisition parameters. Radiology 288(2):407–415
Kim H, Park CM, Lee M et al (2016) Impact of reconstruction algorithms on CT radiomic features of pulmonary tumors: analysis of intra- and inter-reader variability and inter-reconstruction algorithm variability. PLoS One 11:e0164924
Lu L, Ehmke RC, Schwartz LH et al (2016) Assessing agreement between radiomic features computed for multiple CT imaging settings. PLoS One 11:e0166550
Mackin D, Ger R, Dodge C et al (2018) Effect of tube current on computed tomography radiomic features. Sci Rep 8:2354
Zhao YR, Ooijen, Dorrius PMAMD van et al (2014) Comparison of three software systems for semi-automatic volumetry of pulmonary nodules on baseline and follow-up CT examinations. Acta Radiol 55(6):691–698
Kalpathy-Cramer J, Zhao B, Goldgof D et al (2016) A comparison of lung nodule segmentation algorithms: methods and results from a multi-institutional study. J Digit Imaging 29:476–487
Balagurunathan Y, Beers A, Kalpathy-Cramer J et al (2018) Semi-automated pulmonary nodule interval segmentation using the NLST data. Med Phys 45(3):1093–1107
Brochu F. (2019) Increasing shape bias in ImageNet-trained networks using transfer learning and domain-adversarial methods. arXiv preprint arXiv:1907.12892
Armato SG III, McLennan G, Bidaut L et al (2011) The lung image database consortium (LIDC) and image database resource initiative (IDRI): a completed reference database of lung nodules on CT scans. Med Phys 38(2):915–931
Opfer R, Wiemker R (2007) Performance analysis for computer-aided lung nodule detection on LIDC data. Proc. SPIE 6515, Medical imaging 2007: image perception, observer performance, and technology assessment, 65151C
Setio AAA, Traverso A, de Bel T et al (2017) Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: The LUNA16 challenge. Med Image Anal 42:1–13
Reeves AP, Biancardi AM (2011) The Lung Image Database Consortium (LIDC) nodule size report. Release: 2011–10–27–2. http://www.via.cornell.edu/lidc/
Armato SG III, Giger ML, MacMahon H (2001) Automated detection of lung nodules in CT scans: preliminary results. Med Phys 28(8):1552–1561
PH Chen CL, KL Lor, YC Chang, CM Chen (2014) Pulmonary lobe segmentation of 3D thoracic CT images: adaptive rolling ball and vector-based surface deformation. Quantitative CT Imaging of the Lung, Society of Thoracic Radiology. San Antonio, Texas, USA.
Frangi AF, Niessen WJ, Vincken KL, et al (1998) Multiscale vessel enhancement filtering. International Conference on Medical Image Computing and Computer-Assisted Intervention: Springer, pp.130–137
Breiman L (2001) Random Forests. Mach Learn 45:5–32
Golosio B, Masala GL, Piccioli A et al (2009) A novel multithreshold method for nodule detection in lung CT. Med Phys 36(8):3607–3618
Riccardi A, Petkov TS, Ferri G, Masotti M, Campanini R (2011) Computer-aided detection of lung nodules via 3D fast radial transform, scale space representation, and Zernike MIP classification. Medi Phys 38(4):1962–1971
Tan M, Deklerck R, Jansen B, Bister M, Cornelis J (2011) A novel computer-aided lung nodule detection system for CT images. Med Phys 38(10):5630–5645
Tan M, Deklerck R, Cornelis J et al (2013) Phased searching with NEAT in a time-scaled framework: experiments on a computer-aided detection system for lung nodules. Artif Intell Med 59(3):157–167
Chakraborty DP (1989) Maximum likelihood analysis of free-response receiver operating characteristic (FROC) data. Med Phys 16(4):561–567
Chakraborty DP (2013) A brief history of FROC paradigm data analysis. Acad Radiol 20(7):915–919
Funding
This work was supported by the Ministry of Science and Technology, Taiwan, under the grant number MOST107-2221-E-002–074-MY3.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Guarantor
The scientific guarantor of this publication is Chung-Ming Chen, Department of Biomedical Engineering, National Taiwan University.
Conflict of interest
The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.
Statistics and biometry
No complex statistical methods were necessary for this paper.
Informed consent
Written informed consent was waived by the Institutional Review Board.
Ethical approval
Institutional Review Board approval was obtained.
Methodology
• retrospective
• observational
• performed at one institution
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Lin, FY., Chang, YC., Huang, HY. et al. A radiomics approach for lung nodule detection in thoracic CT images based on the dynamic patterns of morphological variation. Eur Radiol 32, 3767–3777 (2022). https://doi.org/10.1007/s00330-021-08456-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00330-021-08456-x