Abstract
The aim of this study is to evaluate the capability of improved artificial adaptive systems and additional novel training methods in order to distinguish between benign and malignant lung nodules in Multi Detector Computed Tomography. A total of 90 nodules belonging to 88 patients are analyzed. A set of adjacent slices representing the lesion selected from the CT Image analysis by the experts are collected and stored in a database. Features extracted by an assembly of adaptive algorithms working in sequence [Active Connection Fusion (ACF): a new set of ANNs for image fusion; J-Net Active Connections Matrix (J-Net): a new ANN for dynamic image segmentation; Population (Pop): a new and fast multidimensional scaling algorithm] are divided into several groups using random or experimental methods to train and test different Artificial Neural Networks. Best results are obtained with Adaptive Learning Quantization (AVQ) and Meta-Consensus, two new supervised ANNs, experts in rapid classification and not sensitive to over fitting. After optimization of the distribution of cases among the training and testing sets the following results are achieved: sensitivity (recognition of malignant nodules) ranging from 93.33% to 100%; specificity (recognition of benign nodules) stable on 95.56% and overall accuracy ranging from 94.44% to 97.78%. These results represent the highest predictive values ever recorded in lung CAD literature. Benchmarking analysis, with advanced mathematical algorithms using simpler approaches, show that complex processing systems, composed of different steps and sub processing systems, are clearly superior and probably needed to reach excellent predictive performances in lung nodule characterization.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The 90 CT volumes were provided by the Department of Radiological Sciences of the University of Rome, “La Sapienza”.
- 2.
Software for extracting ROIs from the original images was set up by Dr. Petritoli and Dr. Terzi (Semeion, Research Center).
- 3.
All the experimentations with Naive Bayes algorithm were executed by Massimiliano Marciano (software engineer at CSI Research and Innovation, via Cesare Pavese 305, Rome, Italy), using Rapid Miner ver. 5.0.010.
References
Aoyama M, Li Q, Kasuragawa S, MacMahon H, Doi K (2002) Automated computerized scheme for distinction between benign and malignant solitary lung nodules on chest images. Med Phys 29:701–708
Aoyama M, Li Q, Katsuragawa S, Li F, Sone S, Doi K (2003a) Computerized scheme for determination of the likelihood measure of malignancy for lung nodules on low-dose CT images. Med Phys 30(3):387–394
Aoyama M, Li Q, Katsuragawa S, Li F, Sone S, Doi K (2003b) Computerized scheme for determination of the likelihood measure of malignancy for pulmonary nodules on low-dose CT images. Med Phys 30:387–394
Blum RS, Liu Z (2006) Multi-sensor image fusion and its applications. Taylor & Francis, London
Buscema M (1998) Meta-Net: the theory of independent judges. In: Buscema M (ed) Substance use & misuse. Marcel Dekker Inc., New York, vol 33(2), pp 439–461
Buscema M and Catzola L (2010) Adaptive vector quantization systems: classic LVQ and new AVQ. Semeion Institute, Rome, IT, Mimeo (in Italian)
Buscema M et al. (2006) Sistemi ACM e imaging diagnostico. Le immagini mediche come matrici attive di connessioni [ACM systems and diagnostic imaging. Medical images as active connections matrices, in Italian] Springer-Verlag, Italy
Buscema M, Breda M and Terzi S (2006a) A feed forward sine based neural network for functional approximation of a waste incinerator emissions. Proceedings of the 8th WSEAS international conference on automatic control, modeling and simulation, Prague
Buscema M, Breda M, Terzi S (2006c) Using sinusoidal modulated weights improve feed-forward neural network performances in classification and functional approximation problems. WSEAS Trans Inf Sci Appl 3(5):885–893
Buscema M, Catzola L, Grossi E (2008) Images as active connection matrixes: the J-Net system. IC-MED Int J Intell Comput Med Sci 2(1):27–53
Buscema M, Terzi S, Tastle W (2010) Meta-consensus: a new meta classifier. In: NAFIPS 2010, 12–14 July 2010, Toronto, Canada, 978-1-4244-7858-6/10 ©2010 IEEE
Diederich S, Wormanns D, Heindel W (2003) Lung cancer screening with low-dose CT. Eur Radiol 45(1):2–7
Diederich S, Wormanns D, Semik M (2002) Screening for early lung cancer with low-dose spiral CT: prevalence in 817 asymptomatic smokers. Radiology 222:773–781
Domingos P, Pazzani M (1997) On the optimality of the simple Bayesian classifier under zero–one loss. Mach Learn 29:103–137
Fischbach F, Knollmann F, Griesshaber V, Freund T, Akkol E, Felix R (2003) Detection of lung nodules by multislice computed tomography: improved detection rate with reduced slice thickness. Eur Radiol 13:2378–2383
Goldin J, Brown M, Petkovska I (2008) Computer-aided diagnosis in lung nodule assessment. J Thorac Imaging 23(2):97–104
Greenlee RT, Murray T, Bolden S, Wingo PA (2000) Cancer statistics, 2000. CA Cancer J Clin 50:7–33
Hand DJ, Yu K (2001) Idiot’s Bayes – not so stupid after all? Int Stat Rev 69(3):385–399, ISSN 0306–7734
Henschke CI, Yankelevitz DF (2008) CT screening for lung cancer: update 2007. Oncologist 13:65–78
Henschke CI, Yankelevitz DF, Libby D et al (2002) CT screening for lung cancer: the first ten years. Cancer J 8:S47–S54
Kawata Y, Niki N, Ohmatsu H, Kakinuma R, Eguchi K, Kaneko M, Moriyama N (1998) Quantitative surface characterization of pulmonary nodules based on thin-section CT images. IEEE Trans Nucl Sci 45:2132–2138
Kohonen T (1995–2001) Self-organizing map, 3rd edn. Springer, Berlin
Kosko B (1992) Neural networks and fuzzy systems. Prentice Hall, New Jersey, pp 39–261
Kunceva LI (2004) Combining pattern classifier. Wiley Interscience, Hoboken, NJ
Kuncheva LI (2004) Combining pattern classifiers: methods and algorithms. Wiley, Hoboken, NJ
Li F, Sone S, Abe H, MacMahon H, Armato AG, Kunio D (2002) Lung cancer missed at low-dose helical CT screening in a general population: comparison of clinical, histopathologic, and imaging findings. Radiology 225:673–683
Li Q (2007) Recent progress in computer-aided diagnosis of lung nodules on thin-section CT. Comput Med Imaging Graph 31(4–5):248–57
Maron ME (1961) Automatic indexing: an experimental inquiry. J ACM (JACM) 8(3):404–417
Massini G, Terzi S and Buscema M (2010) A new method of multidimensional scaling. In: NAFIPS 2010, 12–14 July 2010, Toronto, Canada, 978-1-4244-7858-6/10 ©2010 IEEE
McNitt-Gary MF, Hart EM, Wyckoff N, Sayre JW, Goldin JG, Aberle DR (1999) A pattern classification approach to characterizing solitary pulmonary nodules imaged on high resolution CT: preliminary results. Med Phys 26:880–888
Mozina M, Demsar J, Kattan M, Zupan B (2004) Nomograms for visualization of naive Bayesian classifier. In: Proceedings of PKDD-2004, pp 337–348
Neuralware (1995) Neural computing. A technology for professional II/PLUS and neural networks explorer. NeuralWare Inc., Pittsburg, PA, pp 227–233
Rennie J, Shih L, Teevan J, Tackling K (2003) The poor assumptions of naive Bayes classifiers. In: Proceedings of the twentieth international conference on machine learning (ICML), Washington, DC
Rish I (2001) An empirical study of the naive Bayes classifier. IJCAI 2001 Workshop on empirical methods in artificial intelligence, Seattle, USA
Sammon JW (1969) A non linear mapping for data structure analysis. IDEE Trans Comput C-18(5):401–409
Shiraishi J, Abe H, Engelmann R, Aoyama M, MacMahon H, Doi K (2003) Computer-aided diagnosis to distinguish benign from malignant solitary pulmonary nodules on radiographs: ROC analysis of radiologists’ performance – initial experience. Radiology 227:469–474
Shiraishi J, Li Q, Suzuki K, Li F, Engelmann R, Doi K (2006) Computer-aided diagnostic scheme for the detection of lung nodules on chest radiographs: Localized search method based on anatomical classification. Med Phys 2642:2653
Swensen SJ, Jett JR, Hartman TE, Midthun DE, Sloan JA, Sykes AM, Aughenbaugh GL, Clemens MA (2003) Lung cancer screening with CT: Mayo clinic experience. Radiology 226:756–761
Webb GI, Boughton J, Wang Z (2005) Not so naive Bayes: aggregating one-dependence estimators. Mach Learn 58(1):5–24
Software
Buscema M (1999–2010) Supervised ANNs and organisms, ver. 16.5. Semeion software #12, Rome, Italy
Buscema M (2003–2010) Active connections matrix, ver. 13.5. Semeion software #30, Rome Italy
Buscema M (2008–2010) Meta-Nets, ver. 8.0. Semeion software #44, Rome, Italy
Buscema M (2010) ACF: image fusion, ver.3.5. Semeion software #55, 2010, Rome, Italy
Machine Learning Group (1999–2010) WEKA. Waikato environment for knowledge analysis, ver. 3.6.3, the University of Waikato, Hamilton, New Zealand
Massini G (2007–2009) Population: an fast algorithm for multi-dimensional scaling, ver.3.0. Semeion software #42, Rome, Italy
Rapid I (2001–2010) Rapid Miner, ver. 5.0.010, Dortmund, Germany [http://www.rapidminer.com]
Patents
Buscema M (2003) European patent “An algorithm for recognizing relationships between data of a database and a method for image pattern recognition based on the said algorithm” (Active Connection Matrix –ACM). Owner: Bracco, Semeion. Application no. 03425559.6, deposited August 22, 2003
Buscema M (2004) International patent “An algorithm for recognizing relationships between data of a database and a method for image pattern recognition based on the said algorithm” (Active Connection Matrix – ACM). Owner: Bracco, Semeion. Application n. PCT/EP2004/05182, deposited: August 18, 2004
Buscema M (2007) European patent “Active Connection Matrix J-Net. An algorithm for the processing of images able to highlight the information hidden in the pixels”. Owner: Bracco, Semeion. Application n. 07425419.4, deposited: July 6, 2007
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer Science+Business Media New York
About this chapter
Cite this chapter
Buscema, M., Passariello, R., Grossi, E., Massini, G., Fraioli, F., Serra, G. (2013). J-Net: An Adaptive System for Computer-Aided Diagnosis in Lung Nodule Characterization. In: Tastle, W. (eds) Data Mining Applications Using Artificial Adaptive Systems. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-4223-3_2
Download citation
DOI: https://doi.org/10.1007/978-1-4614-4223-3_2
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4614-4222-6
Online ISBN: 978-1-4614-4223-3
eBook Packages: Computer ScienceComputer Science (R0)