Skip to main content
Log in

Toward Understanding the Size Dependence of Shape Features for Predicting Spiculation in Lung Nodules for Computer-Aided Diagnosis

  • Published:
Journal of Digital Imaging Aims and scope Submit manuscript

Abstract

We analyze the importance of shape features for predicting spiculation ratings assigned by radiologists to lung nodules in computed tomography (CT) scans. Using the Lung Image Database Consortium (LIDC) data and classification models based on decision trees, we demonstrate that the importance of several shape features increases disproportionately relative to other image features with increasing size of the nodule. Our shaped-based classification results show an area under the receiver operating characteristic (ROC) curve of 0.65 when classifying spiculation for small nodules and an area of 0.91 for large nodules, resulting in a 26 % difference in classification performance using shape features. An analysis of the results illustrates that this change in performance is driven by features that measure boundary complexity, which perform well for large nodules but perform relatively poorly and do no better than other features for small nodules. For large nodules, the roughness of the segmented boundary maps well to the semantic concept of spiculation. For small nodules, measuring directly the complexity of hard segmentations does not yield good results for predicting spiculation due to limits imposed by spatial resolution and the uncertainty in boundary location. Therefore, a wider range of features, including shape, texture, and intensity features, are needed to predict spiculation ratings for small nodules. A further implication is that the efficacy of shape features for a particular classifier used to create computer-aided diagnosis systems depends on the distribution of nodule sizes in the training and testing sets, which may not be consistent across different research studies.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

References

  1. American Cancer Society: Cancer facts and figures. Available at http://www.cancer.org/research/cancerfactsfigures/cancerfactsfigures/cancer-facts-figures 2013. Accessed December 2013

  2. Aberle D, Adams A, Berg C, Black W, Clapp J, Fagerstrom R, Gareen I, Gatsonis C, Marcus P, Sicks J: Reduced lung-cancer mortality with low-dose computed tomography screening. N Engl J Med 246:697–722, 2008

    Google Scholar 

  3. Way T, Chan H-P, Hadjiiski L, Sahiner B, Chughtai A, Song TK, Poopat C, Stojanovska J, Frank L, Attili A, Bogot N, Cascade P, Kazerooni EA: Computer-aided diagnosis of lung nodules on CT scans: ROC study of its effects on radiologists’ performance. Acad Radiol 17(3):323–332, 2009

    Article  Google Scholar 

  4. Li F, Li Q, Engelmann R, Aoyama M, Sone S, MacMahon H, Doi K: Improving radiologists’ recommendations with computer-aided diagnosis for management of small nodules detected by CT. Acad Radiol 13:943–950, 2006

    Article  PubMed  Google Scholar 

  5. Armato III, SG, McLennan G, Bidaut L, et al: The lung image database consortium (LIDC) and image database resource initiative (IDRI): a completed reference database of lung nodules on CT scans. Med Phys 38:915–931, 2011

    Article  PubMed Central  PubMed  Google Scholar 

  6. Raicu DS, Varutbangkul E, Cisneros JG, Furst JD, Channin DS, Armato III, SG: Semantic and image content integration for pulmonary nodule interpretation in thoracic computed tomography. SPIE Medical Imaging, San Diego, 2007

    Book  Google Scholar 

  7. Varutbangkul E, Raicu DS, Furst JD: A computer-aided diagnosis framework for pulmonary nodule interpretation in thoracic computed tomography. DePaul CTI Research Symposium, DePaul University, 2007

  8. Muhammad MN, Raicu DS, Furst JD, Varutbangkul E: Texture versus shape analysis for lung nodule similarity in computed tomography studies. Medical Imaging 2008: PACS and Imaging Informatics, 2008

  9. Kim R, Dasovich G, Bhaumik R, Brock R, Furst JD, Raicu DS: An investigation into the relationship between semantic and content based similarity using LIDC. MIR’10 Proceedings of the International Conference on Multimedia Information Retrieval, DOI: 10.1145/1743384.1743417, 185–192, 2010

  10. Varutbangkul E, Mitrovic V, Raicu D, Furst J: Combining boundaries and ratings from multiple observers for predicting lung nodule characteristics. Int Conf Biocomput Bioinform BiomedTechnol 99:82–87, 2008

    Google Scholar 

  11. Horsthemke W, Raicu DS, Furst JD: Evaluation challenges for bridging the semantic gap: Shape Disagreements in the LIDC. Int J Healthcare Information Syst Inform, 2009

  12. Zerhouni EA, Stitik FP, Siegelman SS, Naidich DP, Sagel SS, et al: CT of the pulmonary nodule: a cooperative study. Radiology 160:319–327, 1986

    Article  CAS  PubMed  Google Scholar 

  13. Zhao B, Gamsu G, Ginsburg MS, Jiang L, Schwartz LH: Automatic detection of small lung nodules on CT utilizing a local density maximum algorithm. J Appl Clini Med Phys 4(3):248–260, 2003

    Article  Google Scholar 

  14. Zinovev D, Raicu D, Furst J, Armato III, SG: Predicting radiological panel opinions using a panel of machine learning classifiers. Algorithms 2:1473–1502, 2009

    Article  Google Scholar 

  15. Goldin JG, Brown MS, Petkovska I: Computer-aided diagnosis in lung nodule assessment. J Thoracic Imaging 23(2), 2008

  16. Quang L: Recent progress in computer-aided diagnosis of lung nodules on thin-section CT. Comput Med Imaging Graph 31(4–5):248–257, 2007

    Google Scholar 

  17. Guliato D, Rangayyan RM, Carvalho JD, Santiago SA: Polygonal modeling of contours of breast tumors with the preservation of spicules. IEEE Trans Biomed Eng 55, 2008

  18. Guliato D, de Carvalho JD, Rangayyan RM, Santiago SA: Feature extraction from a signature based on the turning angle function for the classification of breast tumors. J Digit Imaging 21:129–144, 2008

    Article  PubMed Central  PubMed  Google Scholar 

  19. Rangayyan RM, Mguyen TN: Fractal analysis of contours of breast masses in mammograms. J Digit Imaging 20:223–237, 2007

    Article  PubMed Central  PubMed  Google Scholar 

  20. Langlotz CP: RadLex: a new method for indexing online educational materials. RadioGraphics 26:6, 2006

    Article  Google Scholar 

  21. Marwede D, Schulz T, Kahn T: Indexing thoracic CT reports using a preliminary version of a standardized radiological lexicon (RadLex). J Digit Imaging 21:4, 2008

    Article  Google Scholar 

  22. El-Baz A, Beache GM, Gimel’farb G, Sukuki K, Okada K, Elnakib A, Soliman A, Abdollahi B: Computer-aided diagnosis systems for lung cancers: Challenges and methodologies. Int J Biomed Imag, Article ID 9423533, 2013

  23. Furuya K, Murauama S, Soeda H, Murakami J, Ichinose Y, Yabuuchi H, Katsuda Y, Koga M, Masuda K: New classification of small pulmonary nodules by margin characteristics on high-resolution CT. Acta Radiol 40:496–504, 1999

    Article  CAS  PubMed  Google Scholar 

  24. Nakamura K, Yoshida H, Engelmann R: Computerized analysis of the likelihood of malignancy in solitary pulmonary nodules with use of artificial neural networks. Radiology 214:823–830, 2000

    Article  CAS  PubMed  Google Scholar 

  25. Horsthemke WH, Raicu DS, Furst JD: Characterizing pulmonary nodule shape using a boundary-region approach. SPIE Medical Imaging, Orlando, 2009

    Book  Google Scholar 

  26. Horsthemke WH, Raicu DS, Furst JD: Predicting LIDC diagnostic characteristics by combining spatial and diagnostic opinions, medical imaging 2010: Computer-Aided Diagnosis, Proceedings of SPIE, 7624, 2010

  27. Wiemker R, Opfer R, Bulow T, Kabus S, Dharaiya E: Repeatability and noise robustness of spicularity features for computer aided characterization of pulmonary nodules in CT. SPIE Medical Imaging, San Diego, 2008

    Book  Google Scholar 

  28. Weimker R, Bergtholdt M, Dharaiya E, Kabus S, Lee MC: Agreement of CAD features with expert observer ratings for characterization of pulmonary nodules in CT using the LIDC-IDRI database. SPIE Medical Imaging, San Diego, 2009

    Book  Google Scholar 

  29. Kawata Y, Niki N, Ohmatsu H, et al: Classification of pulmonary nodules in thin-section CT images based on shape characterization. Proc Int Conf Imag Process 3(2):528–530, 1997

    Article  Google Scholar 

  30. Kido S, Kuriyama K, Higashiyama M, Kasugai T, Kuroda C: Fractal analysis of small peripheral pulmonary nodules in thin-section CT: evaluation of lung nodule interfaces. J Comput Assist Tomogr 26(4):573–578, 2002

    Article  PubMed  Google Scholar 

  31. El-Baz A, Nitzken M, Vanbogaert E, Gimel’farb G, Faulk R, Abo El-Ghar M: A novel shape-based diagnostic approach for early diagnosis of lung nodules. IEEE Confeon Biomed Imag 137–140, 2011

  32. Huang P-W, Lin P-L, Lee C-H, Kuo CH: A classification system of lung nodules in CT images based on fractional brownian motion model. IEEE International Conference on System Science and Engineering, July 4–6, 2013

  33. Netto SMB, Silva AC, Nunes RA: Analysis of directional patterns of lung nodules in computerized tomography using Getis statistics and their accumulated forms as malignancy and benignity indicators. Pattern Recogn Lett 33:1734–1740, 2012

    Article  Google Scholar 

  34. Way TW, Hadjiiski LM, Sahiner B, Chan H-P, Cascade PN, Kazerooni EA, Bogot N, Zhou C: Computer-aided diagnosis of pulmonary nodules in CT scans: segmentation and classification using 3D active contours. Med Phys 33(7):2323–2337, 2006

    Article  PubMed Central  PubMed  Google Scholar 

  35. Way TW, Sahiner B, Chan H-P, Hadjiiski L, Casscade PN, Chughtai A, Bogot N, Kazerooni E: Computer-aided diagnosis of pulmonary nodules on CT scans: improvement of classification performance with nodule surface features. Med Phys 36(7):3086–3098, 2009

    Article  PubMed Central  PubMed  Google Scholar 

  36. Suzuki K, Li F, Sone S, Doi K: Computer-aided diagnostic scheme for distinction between benign and malignant nodules in thoracic low-dose CT by use of massive training artificial neural network. IEEE Tran Med Imag 24:91138–1150, 2005

    Google Scholar 

  37. Sahiner B, Chan HP, Petrick N, Wei D, Helvie MA, Adler CDD, Goodsitt MM: Image feature selection by genetic algorithm: application to classification of mass and normal breast tissue on mammograms. Med Phys 23:1671–1684, 1996

    Article  CAS  PubMed  Google Scholar 

  38. Dorst F, Smeulders AWM: Length estimators for digitized contours. Computer Vision Graphs Imag Process 40:311–333, 1987

    Article  Google Scholar 

  39. Koplowitz J, Bruckstein AM: Design of perimeter estimators for digitized planar shapes. IEEE Transactions on Pattern Analysis and Machine Intelligence 11(6), June 1989

  40. Montero R, Briciesca E: State of the art of compactness and circularity measures. Int Mathematical Forum 4(27):1305–1335, 2009

    Google Scholar 

  41. Rosenfeld A: Compact figures in digital pictures. IEEE Trans Syst Ma Cybernetics 4(2):221–223, 1974

    Article  Google Scholar 

  42. Lee SC, Wang Y, Lee ET: Compactness measure of digital shapes. Annual Technical and Leadership Workshop, 103–105, 2004

  43. Haralick RM, Shanmugam K, Dinstein I: Textural features for image classification. IEEE Trans Syst Man Cybernet 3(6):610–621, 1973

    Article  Google Scholar 

  44. Andrysiak T, Choras M: Image retrieval based on hierarchical gabor filters. Int J Appl Comput Sci 15(4):471–480, 2005

    Google Scholar 

  45. Breiman L, Friedman J, Olshen R, Stone C: Classification and regression trees. CRC Press, Boca Raton, 1984

    Google Scholar 

  46. Yen SJ, Lee YS: Cluster-based under-sampling approaches for imbalanced data distributions. Expert Syst Applic 36:5718–5727, 2009

    Article  Google Scholar 

  47. Drummond C, Holte RC: C4.5, Class imbalance, and cost sensitivity: why under-sampling beats over-sampling. Workshop on Learning from Imbalanced Data Sets II, 2003

  48. Rahman MM, Davis DN: Addressing the class imbalance problem in medical datasets. Int J Mach Learn Comput 3(2):224–228, 2013

    Article  Google Scholar 

  49. Bogot NR, Kazerooni EA, Kelly AM, Quint LE, Desjardins B, Nan B: Interobserver and intraobserver variability in the assessment of pulmonary nodule size on CT using film and computer display methods. Acad Radiol 12:948–956, 2005

    Article  PubMed  Google Scholar 

  50. Chen H, Xu Y, Ma Y, Ma B: Neural network ensemble-based computer-aided diagnosis for differentiation of lung nodules on CT images. Acad Radiol 17:595–602, 2010

    Article  PubMed  Google Scholar 

  51. Shie, C-Z, Zhao Q, Luo L-P, He J-X: Size of solitary pulmonary nodule was the risk factor for malignancy. J Thorac Dis 6(6), 2014

  52. Xu DM, van Klaveren RJ, de Bock GH, Leusveld A, Zhao Y, Wang Y, Vliegenthart R, de Koning HJ, Scholten ET, Verschakelen J, Prokop M, Oudkerk M: Limited value of shape, margin, and CT density in the discrimination between benign and malignant screen detected solid pulmonary nodules of the NELSON trial. Eur J Radiol 68(2):347–352, 2007

    Article  PubMed  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ron Niehaus.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Niehaus, R., Stan Raicu, D., Furst, J. et al. Toward Understanding the Size Dependence of Shape Features for Predicting Spiculation in Lung Nodules for Computer-Aided Diagnosis. J Digit Imaging 28, 704–717 (2015). https://doi.org/10.1007/s10278-015-9774-8

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10278-015-9774-8

Keywords

Navigation