Computer aided detection of pulmonary nodules using genetic programming

Computer-aided Detection of Pulmonary Nodules using Genetic Programming

Wook-Jin Choi and Tae-Sun Choi

2

Contents

• Introduction• Lung Segmentation based on 3D Approach• Nodule Candidates Detection and Feature

Extraction• Genetic Programming Based Classification• Experimental Results• Conclusions• References

3

Introduction• Pulmonary nodule detection is attractive applications of computer-

aided detection (CAD) because lung cancer is the leading cause of cancer deaths.

• If lung cancer detected in early phase, the 3-year survival rate is more than 80%.

• Recently, researchers have developed a number of CAD methods for lung nodules to aid radiologists in identifying nodule candidates from CT images.

• Current CT technology allows for near isotropic, submillimeter reso-lution acquisition of the complete chest in a single breath hold.

• These thin-slice chest scans have become indispensable in thoracic radiology, but have also substantially increased the data load for radiologists.

• Automating the analysis of such data is, therefore, a necessity and this has created a rapidly developing research area in medical imaging.

4

Related Works• Template matching methods

– Genetic Algorithm Template Matching [10]– 3D Template Matching [11]

• Model based methods– Patient-specific models [5]– Surface normal overlap model [7]

• Machine learning techniques– Neural network [6]– Fuzzy c-means clustering [9]

• Digital filtering – Quantized convergence index filter [8]– Iris filter [13]

• Statistical analysis [12]

5

Proposed Algorithm

Flow chart of Pulmonary nodule detection

6

Lung Segmentation based on 3D Approach

7


• Select adaptive threshold value at every slice in the CT im-age sequence using diagonal intensity histogram [4].

• The CT images are divided into background area(body) and foreground area(air or lung) as shown below.

Original CT image and converted CT image with thresh-old

8


Extracted lung region using 3D connected component labeling and contour corrected lung region (containing wall side nodule)

• Segment lung region and remove the rim (outer part of the body).

• Correct the contour of the lung volume (correct excluded wall side nodule).

9

Nodule Candidates Detection and Feature Ex-traction

10

ROI Extraction

6-stepped ROI and ex-tracted nodule candidates

• Adaptive multiple thresh-olding method.– the traditional multiple

thresholding method makes many steps of grey levels.

– We calculate the adaptive threshold value using diag-onal histogram at every slice of lung volume.

– This value is base threshold value for multiple threshold-ing.

– We make additional five threshold values which are base threshold + 50, -50, -100, -150 and -200.

11

Nodule Candidates Detec-tion• We can remove the vessels and

noise in the lung volume using rule based classifier.

• Vessel RemovingThe vessel is classified by volume elongation factor and compact-ness.– volume is extremely bigger than

nodule – longer than nodule– not compact object.

• Noise removing– radius of ROI is smaller than 3mm or

bigger than 30mm.

• Remaining ROIs are nodule candi-dates

6-stepped ROI and ex-tracted nodule candidates

12

Feature Extraction• 3D geometric features

– Volume– elongation factor– Compactness– approximated radius.

• 2D pixel based features.– Use median slice of nodule candidates (area of the median slice is the

largest)– To extract 2D texture feature, we normalize the image size of nodule

candidates. – 3 types of nodule sizes and then extract the features.

• < 5mm : the size of image matrix is 8x8.• 5mm ~10mm : the size of image matrix is 16x16.• > 10 mm : the size of image matrix is 32x32

– extract 14 features from the image matrix.• mean, variance, skewness, kurtosis, area, radius and 8 biggest eigenvalues.

13

Feature ExtractionIndex Feature

1 Z position

2 Mean

3 Variance

4 Skewness

5 Kurtosis

6 Area

7 Radius

8 Perimeter

9 Compactness

10~17 Largest Eigenvalue 1~8

18 X centroid

19 Y centroid

20 Z centroid

21 Width

22 Height

23 Depth

24 Size

14

Genetic Programming Based Classification

15


• Genetic Programming (GP) – an evolutionary optimization technique [14].

• The basic structure of GP is very similar to Genetic Algorithm(GA).

• The chromosome– GA : variable (binary digit)– GP : program (tree or graph)

16


A function represented as a tree structure

17


• Our goal of GP evolution is to reduce false posi-tive (FP) and increase true positive (TP).

• In the proposed scheme, an optimized classifier is carried out using combination of features and random constant values.

• GP optimally selects adequate features from all extracted features and combines the selected features with mathematical operators.

• The GP generates individual classifiers and those are evaluated by fitness function.

• The result of GP can convert complex input fea-tures to simple value.

18


• GP chromosome – The terminal set - The elements of feature vector extracted

from nodule candidate images and randomly generated con-stants with in the range 0,1.

– The function set - Four standard arithmetic operator namely plus, minus, multiply and division and additional mathematical operators log, exp, abs, sin and cos.(All operators in the func-tion set are protected to avoid exception)

• GP evolves combination of the terminal set and function set.

19


• Fitness Function – evaluate every individuals in GP generation– True positive rate (TPR)

– Specificity (SPC)• SPC is the value subtracted from 1 to FPR and also called true negative

rate(TNR).

– Area under the ROC curve (Az)• ROC curve is plotted between TP and FP for different threshold values.• Az is area under the ROC curve and a good measure of classifier perfor-

mance in different condition.

– Fitness Function

TPTPR

TP FN

1 1TN FP

SPC FPRTN FP FP TN

* *f TPR SPC Az

20


Objective To evolve maximum fitness

Selection Generational

Population Size 300

Generation Size 80

Initial Tree Depth

Limit6

Initial population Ramped half and half

GP Operators prob Variable ratio of crossover mutation is used

Sampling Tournament

Survival mechanism Keep the best individuals

Real max. tree level 30

Genetic Programming parameter

21


• Examples of GP– minus(minus(P(21,:),exp(P(23,:))),minus(mypower(mylog(plus(

times(P(14,:),minus(P(23,:),mypower(mylog(plus(times(P(12,:),minus(P(11,:),mypower(P(13,:),P(13,:)))),P(22,:))),P(13,:)))),minus(P(20,:),cos(exp(P(7,:)))))),mypower(exp(P(7,:)),P(7,:))),times(minus(minus(mypower(exp(P(23,:)),P(7,:)),P(11,:)),times(exp(P(23,:)),P(12,:))),P(11,:))))

– minus(minus(plus(P(4,:),P(7,:)),sin(minus(P(7,:),mypower(P(24,:),plus(minus(P(7,:),P(11,:)),mypower(P(15,:),P(7,:))))))),mypower(mypower(mypower(P(24,:),exp(P(11,:))),minus(plus(P(10,:),mypower(plus(minus(P(13,:),sin(exp(P(12,:)))),mypower(P(24,:),plus(P(13,:),P(4,:)))),plus(plus(minus(0.35089,0.35089),P(3,:)),P(7,:)))),P(11,:))),minus(plus(P(10,:),minus(P(4,:),plus(P(10,:),P(7,:)))),mypower(minus(plus(P(4,:),exp(P(2,:))),minus(P(9,:),P(4,:))),P(11,:)))))

22

Experimental Results• Lung Image Database Consortium (LIDC) database [15]

– to evaluate the performance of the proposed method.– The LIDC is developing a publicly available database of thoracic com-

puted tomography (CT) scans as a medical imaging research resource to promote the development of computer-aided detection or character-ization of pulmonary Nodules.

– The database is separated into 84 cases, each containing around 100-400 Digital Imaging and Communication (DICOM) images and an XML data file containing the physician annotations

• We applied our method to 32 scans consisting of 153 nod-ules and 7528 slices. The pixel size in the database ranged from 0.65 to 0.75 mm and the reconstruction interval ranged.

• The half of dataset(16 scans) is used for training and an-other half of dataset(another 16 scans) is used for testing the classifier.

23

Experimental Results

(a) (b)

The result of pulmonary nodule detection: (a) 43rd slice, (b) vol-ume rendering

24


Data set TPR FPR Az

learn 93.33% 0.127 0.934

test 91.67% 0.138 0.897

all 92.31% 0.133 0.912

The results of GP based classifier

25


ROC curves of GP based classifier with respect to three datasets

26

Conclusion• We have proposed a novel pulmonary nodule detection algorithm

in CT images.• Lung region is segmented using adaptive thresholding and voxel

labeling based method.• Then nodule candidates are detected using adaptive multiple

thresholding and rule based classifier with 3D geometric features.• Next, 3D and 2D features are extracted from the detected nodule

candidates.• Finally, the extracted features are optimized and then classified

into nodule and non-nodule using GP.• We applied proposed algorithm to the LIDC database of NCI.• This method extremely reduced FP rate.• The FPs per scan is only 6.5 with more than 90% sensitivity.• The results show the superiority of the proposed method.

27

References• [1] Ahmedin Jemal, Rebecca Siegel, ElizabethWard, Yongping Hao, Jiaquan Xu, and

Michael J Thun, “Cancerstatistics, 2009,” CA Cancer J Clin, vol. 59, no. 4, pp. 225–49, Jan 2009.

• [2] K-W Jung, Y-J Won, S Park, H-J Kong, J Sung, H-R Shin, E-Cl Park, and J S Lee, “Cancer statistics in korea: incidence, mortality and survival in 2005,” J Korean Med Sci, vol. 24, no. 6, pp. 995–1003, Dec 2009.

• [3] Qiang Li, “Recent progress in computer-aided diagnosis of lung nodules on thin-section ct.,” Comput Med Imaging Graph, vol. 31, no. 4-5, pp. 248–257, 2007.

• [4] S G Armato, M L Giger, C J Moran, J T Blackburn, K Doi, and H MacMahon, “Com-puterized detection of pulmonary nodules on ct scans,” Radiographics, vol. 19, no. 5, pp. 1303–11, Jan 1999.

• [5] M Brown, M McNitt-Gray, J Goldin, R Suh, J Sayre, and D Aberle, “Patient-specific models for lung nodule detection and surveillance in ct images,” IEEE TMI, vol. 20, no. 12, pp. 1242 – 1250, Dec 2001.

• [6] K Suzuki, SG Armato III, F Li, S Sone, and K Doi, “Massive training artificial neural network (mtann) for reduction of false positives in computerized detection of lung nodules in low-dose computed tomography,” Medical physics, vol. 30, pp. 1602, 2003.

• [7] D Paik, C Beaulieu, G Rubin, B Acar, R Jeffrey, J Yee, J Dey, and S Napel, “Surface normal overlap: a computer-aided detection algorithm with application to colonic polyps and lung nodules in helical ct,” IEEE TMI, vol. 23, no. 6, pp. 661 – 675, Jun 2004.

• [8] Sumiaki Matsumoto, Harold L Kundel, James C Gee, Warren B Gefter, and Hiroto Hatabu, “Pulmonary nodule detection in ct images with quantized convergence index filter.,” Med Image Anal, vol. 10, no. 3, pp. 343–352, Jun 2006.

28

References• [9] N Memarian, J Alirezaie, and P Babyn, “Computerized detection of lung nodules

with an enhanced false positive reduction scheme,” IEEE ICIP 2006, pp. 1921 –1924, Sep 2006.

• [10] Jamshid Dehmeshki, Xujiong Ye, Xinyu Lin, Manlio Valdivieso, and Hamdan Amin, “Automated detection of lung nodules in ct images using shape-based genetic algo-rithm.,” Comput Med Imaging Graph, vol. 31, no. 6, pp. 408–417, Sep 2007.

• [11] Onur Osman, Serhat Ozekes, and Osman N Ucan, “Lung nodule diagnosis using 3d template matching.,” Comput Biol Med, vol. 37, no. 8, pp. 1167–1172, Aug 2007.

• [12] A El-Baz, G Gimel’farb, R Falk, and M Abo El-Ghar, “Automatic analysis of 3d low dose ct images for early diagnosis of lung cancer,” Pattern Recognition, vol. 42, no. 6, pp. 1041–1051, Jan 2009.

• [13] JJ Su´arez-Cuenca, PG Tahoces, M Souto, MJ Lado, M Remy-Jardin, J Remy, and J Jos´e Vidal, “Application of the iris filter for automatic detection of pulmonary nodules on computed tomography images,” Computers in Biology and Medicine, 2009.

• [14] J Koza, “Genetic programming: On the programming of computers by means of natural selection,” The MIT Press, Jan 1992.

• [15] S G Armato, G McLennan, M F McNitt-Gray, C R Meyer, D Yankelevitz, D R Aberle, C I Henschke, E A Hoffman, E A Kazerooni, H MacMahon, A P Reeves, B Y Croft, L P Clarke, and Lung Image Database Consortium Research Group, “Lung image data-base consortium: developing a resource for the medical imaging research commu-nity.,” Radiology, vol. 232, no. 3, pp. 739–748, Sep 2004.

29

Thank You

Technology

Computer aided detection of pulmonary nodules using genetic programming