Caries Detection2
Caries Detection2
Caries Detection2
DETECTION
Faculty of Computing
Universiti Teknologi Malaysia
JUNE 2015
iii
Dedicated to my beloved parents and family, whom without their love and support this
research would have never been completed.
iv
ACKNOWLEDGEMENT
I would also like to say special thanks my co-supervisors Dr. Ismail Bin
Mat Amin and Dr. Nor Ashikin Bte Sharif, whose precious guidance, support and
encouragement were pivotal in establishing my self-confidence in this endeavor.
I would like to thank the staff of Universiti Teknologi Malaysia, and especially
the Faculty of Computing, for their kind cooperation.
ABSTRACT
Caries detection system is important for dental disease diagnosis and treatment.
It can be identified using X-ray imaging. The X-ray image contains interest point of
dental to get the teeth information according to specific diagnostic intention. The
Region of Interest (ROI) includes the caries area on tooth surface. The imaging
challenges like noise, intensity inhomogeneities and low contrast causes the difficulty
for identifying correctly the ROI in dental images. According to the recent studies,
among all medical image segmentation methods, level set has the best segmentation
accuracy. However, there are several components in the level set that need to be
enhanced to determine the exact boundary to separate the ROI. The signed force
function to control the direction of level set evaluation process, speed function to
control the speed of movement and Initial Contour (IC) generation to obtain a more
accurate ROI require an enhancement for the better accuracy. In this research, a new
enhancement of segmentation method has been proposed based on finding an accurate
outcome. The method includes two phases: IC generation and intelligent level set
segmentation. In addition, caries detection process is performed with new detection
method. To generate the IC for dental X- ray images, a new local IC selection for
level set method is proposed. Statistical and morphological information of image is
extracted to establish a technique that is able to find a suitable IC. In the second phase,
statistical information of the pixels inside and outside the generated contour and linear
motion filtering is used to construct the region-based signed force function to provide
more stabilisation to proposed method. Furthermore, 31 features of image are extracted
to train the neural network and to generate proper speed function parameter. The results
of proposed method provide the high accuracy and efficiency in the process of getting
teeth boarder. The next process is to detect from the segmented images. The research
also proposed a new method using integral projection and feature map for every single
tooth to obtain the information of caries area. The achieved overall performance of
proposed segmentation method is evaluated at 120 periapical dental radiograph (X-
ray), with 90% accuracy rate. In addition, the caries detection accuracy rate on 155
segmented images is 98%.
vi
ABSTRAK
Sistem pengecaman karies adalah penting bagi proses dianogsis dan rawatan
penyakit pergigian. Ini boleh dikenalpasti melalui imej X-ray. Imej X-ray
mengandungi maklumat penting untuk mendapatkan informasi berkaitan gigi bagi
tujuan diagnostik secara khusus. Kawasan Kepentingan (ROI) mengandungi maklumat
kawasan permukaan gigi tersebut. Cabaran yang perlu ditangani adalah kekotoran
imej, kedalaman cahaya dan rendah kontras menyebabkan ROI bagi imej X-ray gigi
tidak dapat dikenal pasti dengan tepat. Berdasarkan kepada kajian semasa terhadap
semua kaedah segmentasi imej perubatan, kaedah set tahap adalah kaedah yang
memberi nilai ketepatan yang baik. Walaupun begitu, masih terdapat komponen dalam
kaedah segmantasi set tahap memerlukan peningkatan kaedah segmentasi dengan
menentukan sempadan yang tepat untuk memisahkan ROI. Fungsi daya adalah untuk
mengawal arah proses pengujian set tahap, fungsi kelajuan adalah untuk mengawal
kadar kecepatan pengembangan dan penjanaan sempadan asas untuk mendapatkan
ROI yang lebih tepat memerlukan penambahbaikan untuk mendapatkan ketepatan
segmentasi yang lebih tepat. Dalam kajian ini, kaedah segmentasi yang baru telah
dicadangkan berdasarkan kaedah set tahap untuk mendapatkan hasil yang lebih tepat.
Kaedah tersebut mempunyai dua fasa: iaitu penghasilan Kontur Awalan (IC) dan
kepintaran segmentasi imej berlandaskan kaedah set tahap. Di samping itu, proses
pengesanan karies akan dilaksanakan dengan kaedah pengesanan yang baru. Bagi
menjana IC pada imej, satu kaedah IC yang baru untuk kaedah set tahap telah
dicadangkan. Maklumat statistik dan morfologi imej diekstrak untuk menghasilkan
satu teknik yang boleh mencari IC yang sesuai. Pada fasa kedua, maklumat statistik
untuk nilai piksel dalam dan luar kontur yang terhasil dan penggunaan penapisan
gerakan linear digunakan untuk menjana fungsi daya berdasarkan kawasan bagi
mengawal dan menyediakan lebih kestabilan terhadap kaedah yang dicadangkan.
Selain itu, 31 ciri-ciri imej yang telah diekstrak untuk melatih rangkaian neural dan
menjana parameter fungsi kelajuan yang bersesuaian. Hasil kajian menunjukkan
kaedah yang telah dicadangkan memberi nilai ketepatan yang tinggi dan efisien dalam
proses mendapatkan sempadan gigi. Proses seterusnya adalah untuk mengesan karies
daripada imej yang telah disegmentasi. Kajian ini turut mencadangkan kaedah baru
menggunakan kaedah unjuran integral dan peta sifat yang telah dibangunkan untuk
setiap gigi bagi mendapatkan maklumat kawasan karies. Prestasi keseluruhan yang
dicapai daripada kaedah segmentasi yang dicadangkan dinilai dengan 120 radiograf
gigi periapical (X-ray), nilai ketepatan 90%. Di samping itu, pengesanan karies untuk
155 imej yang disegmentasi adalah pada nilai ketepatan 98%.
vii
TABLE OF CONTENTS
DECLARATION ii
DEDICATION iii
ACKNOWLEDGEMENT iv
ABSTRACT v
ABSTRAK vi
TABLE OF CONTENTS vii
LIST OF TABLES xi
LIST OF FIGURES xii
LIST OF ABBREVIATIONS xv
LIST OF APPENDICES xvii
1 INTRODUCTION 1
1.1 Introduction 1
1.2 Problem Background 2
1.3 Problem Statement 8
1.4 Research Goal 10
1.5 Research Objectives 10
1.6 Research Scope 10
1.7 Significance of Study 11
1.8 Organization of The Thesis 12
2 LITERATURE REVIEW 14
2.1 Introduction 14
2.2 Dental Imaging 15
2.3 Caries Categorization 20
2.4 Image Enhancement 23
2.5 Segmentation 26
2.6 Dental Radiograph Segmentation Approaches 26
viii
2.6.1 Thresholding 28
2.6.1.1 Adaptive Thresholding 30
2.6.1.2 Iterative Thresholding 31
2.6.2 K-Means Clustering 34
2.6.3 Region Growing 36
2.6.4 Active Contour 39
2.6.5 Level Set 44
2.6.6 Integral projection 50
2.7 Feature Extraction 52
2.7.1 Gray Level Co-occurrence Matrix 53
2.7.2 Histograms of Oriented Gradients 54
2.7.3 Boundary Corner Detection 55
2.7.4 Geometric Moment 57
2.8 Morphological Operations 59
2.9 Linear Motion Filtering 60
2.10 Discussion 61
2.11 Summary 66
3 RESEARCH METHODOLOGY 67
3.1 Introduction 67
3.2 Research Framework 68
3.3 Data Collection 71
3.4 Pre-processing 71
3.5 Segmentation 72
3.5.1 Initial Contour Generation 72
3.5.2 Intelligent Level Set 74
3.6 Caries Detection 75
3.7 Evaluation 76
3.8 Summary 77
4 DATASET DEVELOPEMENT 78
4.1 Introduction 78
4.2 Dataset Development 79
4.3 Image Acquisition 80
4.3.1 Equipment Specification 80
4.3.2 Software Specification 82
4.3.3 Data Acquisition Process 82
4.3.4 Data Specification 83
ix
4.4 Pre-Processing 86
4.5 Summary 87
REFERENCES 163
Appendices A B 172 188
xi
LIST OF TABLES
LIST OF FIGURES
LIST OF ABBREVIATIONS
A Area
ANN Artificial Neural Network
AR Abnormal Region
BR Background Region
BPNN BackPropagation Neural Networks
C Constant
CAD Computer-Aided Design
CT Computed Tomography
E Energy
ESF Edge Stopping Function
FP False Positives
GLCM Grey Level Co-occurrence Matrix
HOG Histogram of Oriented Gradient
IC Initial Contour
ILS Intelligent Level Set
IP Integral Projection
LS Level Set
LT Local Threshold
MRI Magnetic Resonance Imaging
MRIBC Morphological Region-Based Initial Contour
NR Normal Region
PAR Potentially Abnormal Region
PPV Predefined Pixels Value
R Region
ROI Region of Interest
SDF Signed Distance Function
SFF Signed Force Function
xvi
LIST OF APPENDICES
INTRODUCTION
1.1 Introduction
of population.
Dental X-ray images are an important imaging examination, which can provide
the detail of teeth and gums to dentist for early diagnosis and treatment purposes. It
is considerable that failure in disease diagnosis by dentist or any specialist will cause
damage to the patient health. However, it is normal that human makes mistakes in
disease diagnosis even dentists or specialist, but what is the solution to reduce these
mistakes to minimum or even zero? To overcome these problems, development of
computer-aided diagnosis and caries detection algorithm has turn into a priority to
help the dentists and specialists to make a better and faster decision on diagnosis and
furthermore treatment.
it shows the size, location and condition of teeth. In addition, they are able to detect the
presence of cavities before they appear on the tooths surface. The X-rays are cheaper
than other imaging techniques and easy to access.
The teeth have many structures which can be distinguished by the features and
textures appeared on the images. Structures which are dense (such as metal restoration
or silver fillings) will appear white on film because of blocking the most of photons.
The structures of empty area which contains just air, will appear by black on a film,
and the other portions that will appear as shades of gray, which contain tissue, teeth
and fluid. Dentists are now able to perform early diagnosis and treatment by observing
the non-intrusive estimates such as the roughness and tooth surface which shows the
texture and moreover, opacification and changes in tooth surface color (Roberson
et al., 2006). Furthermore, investigation or examination using human eye which has
low sensitivity rate has less accuracy and the chance of missing some caries or miss
treatment will occur (Olsen et al., 2009). Moreover, detecting of some cavities in early
stages cannot be seen by visual examination and even difficult to detect by human
vision system in X-ray images.
The location, affection of hard tissues, etiology and rate of caries progress are
the factors for dental caries classification (Sonis, 2003). Mostly the caries regions on
tooth surface is not visible to human eyes and it is difficult to recognise it. Figure 1.1
shows barely visible caries on tooth over time.
4
(a) (b)
Figure 1.1: Caries area on dental X-ray image (a) the X-ray without caries. (b) the X-
ray image which has been taken some months later, and can detect the cavities between
the teeth, which is not visible by visual examination.
Figure 1.2: Segmentation of bitewing X-ray image and extracted ROI of each tooth
with rectangle boundary around it.
Most researches on segmentation have been done within the past ten years as
the computational ability of computers has increased. There is no uniquely superior
technique, as each application presents its own specific challenges. It would be foolish
at best to expect one algorithm alone to be successful at segmentation in any affected
problem areas. This is especially true when the objects under consideration are natural
(biological), rather than man-made, since: 1) biological objects exist in many shapes,
sizes, colors and textures even within a single species, and 2) many different species
resemble one another. Segmentation of medical images is more challenging problem
due to some facts such as large variety in topologies, the complexity of structures
and imaging problems like poor image qualities, low contrast, noise, several types of
6
However, working with X-ray images has some difficulties like noise, intensity
in homogeneities and low contrast between certain tissues. The visible tissue
around the teeth is gum which has similar intensity that caused more difficulties on
segmentation process. Also this type of image is gray level, thus using the color
feature in not possible in processing task. Noise in medical X-ray images has a
number of origins, but the most fundamental is from the X-ray source itself. This
type of noise is called Quantum noise, in reference to the discrete nature of the X-
ray photons producing it (Sprawls, 1995). That is why segmentation on some teeth
boundary has some complexity. There are many different factors that causes the
intensity inhomogeneity in each images which one of the most common problem is
the source of image producing device which depends on the variations of imaging
situation (Chunming et al., 2008).
Recent work on dental image segmentation is (Lin et al., 2014) which has
been performed based on local singularity analysis. They used connected component
analysis and Otsus thresholding to recognize each tooth. Moreover, they utilized
snake boundary tracking for tooth delineation and morphological operation. The
experimental result on 28 periapical dental X-ray images which consist of 75 useful
teeth shows the accuracy of True Positive (TP) by 0.8959 and False Positive (FP) by
0.0093.
Level-set methods (Osher and Sethian, 1988) have been more interesting
to researchers from different areas (Deng and Tsui, 2002; Nilsson and Heyden,
2003; Jeon et al., 2005). The advantage of this method is that the all level-sets
represent a nice extraction of regions and boundaries without uses of complex data
structures. However, level-set function is limited to separation of two regions. And
if it is more than two regions, the level-set method loses the parts of its capability.
Moreover, the result depends on initial contour placement (Shrimali et al., 2009). The
researchers (Shuo et al., 2006, 2007) used variational level-set method through the use
of SVM to train and find the initial contour point. However, their work is not accurate
and it is not an automatic method, and needs human involvement and more over it is
time consuming process.
Recently many studies have been done to improve the image segmentation
using level set method. Taheri et al. (2010) used intensity thresholding to define the
speed function in level set for 3D tumor image segmentation. Proposed level set based
segmentation method is efficient while the image is clear enough and boundaries of
tumor and non-tumor regions is visible. In case, the method is not suitable for noisy
images such as dental X-ray images. Zhang et al. (2010) proposed a region based
active contours method. They utilized the gussian filtering for regularizing the level
set segmentation method and reduced re-initialization of traditional level set method.
However, method still depends on selection of initial contour and poor selection of
initial contour will cause the result of segmentation.
Therefore, the accuracy of every process in such a dental image analysis are
important for clinical diagnosis applications to reduce the risk of wrong diagnosis and
speed up the treatment process. In this research the X-ray images are analyzed in order
to detect the caries which are not visible by human visual inspection. The segmentation
is challenging process in this research which requires improvement to obtain the high
accuracy in result of caries detection.
Among well-known contributions on this problem, level set based method has
shown effective performance on medical and dental images segmentation. However,
the accuracy of the image segmentation process need to be improve for dental X-ray
images. It is because the images contain noises from imaging equipment and reflection
of dental works and moreover the poor boundaries between tissues. Therefore the
manual process of initial contour selection in level set method need to intelligently
automate to obtain the most suitable initial contour points for getting the accurate
segmentation results. Beside that, the speed function and signed force function
9
In most of the image segmentations using level set function, the contour level
zero or initial contour has been chosen manually for all images (Gao and Chae, 2010;
Shuo et al., 2006, 2007), and segmentation process is done without selecting the proper
initial contour which affect the result of segmentation. Selection of proper initial
contours manually for each image is also a time consuming process. Images that
contain noise or images that need to segment the local regions will fail in segmentation
process by using this method.
One of the most important and yet largely unsolved issues in level set
segmentation framework is parameter selection for speed function and signed force
function (Taheri et al., 2010). Usually, parameter of speed function fixed beforehand
by the developer of algorithm and that will result in the best possible segmentation
for images (Li et al., 2011; Chunming et al., 2011; Shuo et al., 2007). Inappropriate
choice of parameters may result in unsatisfactory segmentation and user may have
to spend a significant amount of time correcting the segmentation. Furthermore, the
proper signed force function will speed up the method and increase the accuracy of
segmentation of images with poor boundaries.
Detection of caries in each segmented tooth requires to extract the new features
of each tooth surface and identify the caries and non caries area. Develop a feature
map of each tooth surface will speed up the detection process
10
To propose a new scheme to generate the appropriate initial contour for each
images which causes more accurate result in dental image segmentation.
To enhance the level set method by using morphological concept and artificial
neural network for define the speed function parameter selection with new signed
force function parameter to increase the segmentation performance.
The research objectives are achieved by identifying the problem scope which
covers the following aspects:
and level set based method by utilizing motion filtering for signed force function
and moreover neural network and feature extraction methods to generate the
speed function parameter for dental X-ray image segmentation.
Dental diseases has high risk of affection in the globe and mostly in adult
population. Worldwide WHO (World Health Organization) studies reports that the
main purpose of tooth loss and most of dental disease causes by dental caries which is
involving large population in world. Dental caries is one of the main concerns in dental
diagnosis and treatment community which more than 90% of all the adults have dental
caries (WHO, 2012). In Malaysia also more than 80% of adults have dental disease
and suffering from it.
The study shall contribute to four area of digital image processing; the image
enhancement, the image segmentation, feature extraction and finally classification
process. Segmentation of such a dental X-ray images compare to other medical images
due to its difficulties is more challenging procedure. The segmentation must be able to
eliminate the background of X-ray image of teeth and achieve the individual tooth. It
12
method. This chapter also discusses all the proposed techniques involved in
implementation of proposed methodology on dental X-ray image segmentation and
caries detection approach.
Chapter 4 presents the dataset development of this study and describes pre-
processing process for obtained images. This chapter discusses the specification of
dataset used in this study and how to make a better quality out of reduced image quality
to achieve reliable results.
Chapter 7 demonstrates the experimental results for current dental X-ray image
dataset. The results of each periapical radiograph image segmentation are presented
with explanation. Furthermore, the popular and recent research in dental image
segmentation implemented and compared with this work to evaluate the performance
and accuracy of this research. This chapter also provides comparison between
proposed method with other famous methods in dental X-ray image segmentation with
current dataset.
Finally, chapter 8 concludes the major achievements drawn from this research
and future directions are recommended.
REFERENCES
Chen, Y., Thiruvenkadam, S., Tagare, H. D., Huang, F., Wilson, D. and Geiser, E. A.
(2001). On the incorporation of shape priors into geometric active contours. In
Variational and Level Set Methods in Computer Vision, 2001. Proceedings. IEEE
Workshop on. IEEE, 145152.
Choras, R. S. (2007). Image feature extraction techniques and their applications
for CBIR and biometrics systems. International journal of biology and biomedical
engineering. 1(1), 616.
Chunming, L., Huang, R., Ding, Z., Gatenby, C., Metaxas, D. and Gore, J. (2008).
A variational level set approach to segmentation and bias correction of images
with intensity inhomogeneity. In Medical Image Computing and Computer-Assisted
InterventionMICCAI 2008. (pp. 10831091). Springer.
Chunming, L., Huang, R., Ding, Z., Gatenby, J., Metaxas, D. N. and Gore, J. C.
(2011). A level set method for image segmentation in the presence of intensity
inhomogeneities with application to MRI. Image Processing, IEEE Transactions on.
20(7), 20072016.
Ciofolo, C. (2004). Atlas-based segmentation using level sets and fuzzy labels. In
Medical Image Computing and Computer-Assisted InterventionMICCAI 2004. (pp.
310317). Springer.
Clarke, L., Velthuizen, R., Camacho, M., Heine, J., Vaidyanathan, M., Hall, L.,
Thatcher, R. and Silbiger, M. (1995). MRI segmentation: methods and applications.
Magnetic resonance imaging. 13(3), 343368.
Dalal, N. and Triggs, B. (2005). Histograms of oriented gradients for human detection.
In Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer
Society Conference on, vol. 1. IEEE, 886893.
Deng, J. and Tsui, H.-T. (2002). A fast level set method for segmentation of low
contrast noisy biomedical images. Pattern Recognition Letters. 23(1), 161169.
Dougherty, G. (2009). Digital image processing for medical applications. Cambridge
University Press.
Ehsanirad, A. (2010). Plant classification based on leaf recognition. International
Journal of Computer Science and Information Security. 8(4), 7881.
Elyasi, A., Ganjdanesh, Y., Kangarloo, K. and Hossini, M. (2011). Level set
segmentation method in cancers cells images. Journal of American Science. 7(2),
196204.
Eyad Haj, S., Nassar, D. E. M. and Ammar, H. H. (2006). Image segmentation for
automated dental identification. In Electronic Imaging 2006. International Society
165
hierarchical image segmentation with level set and additive operator splitting.
Pattern Recognition Letters. 26(10), 14611469.
Kass, M., Witkin, A. and Terzopoulos, D. (1988). Snakes: Active contour models.
International journal of computer vision. 1(4), 321331.
Keyes, L. and Winstanley, A. (2001). Using moment invariants for classifying shapes
on large-scale maps. Computers, Environment and Urban Systems. 25(1), 119130.
Kimmel, R. and Bruckstein, A. M. (2003). Regularized Laplacian zero crossings as
optimal edge integrators. International Journal of Computer Vision. 53(3), 225243.
Krol, D. M. (2003). Dental caries, oral health, and pediatricians. Current Problems in
Pediatric and Adolescent Health Care. 33(8), 253 270. ISSN 1538-5442.
Kronfeld, T., Brunner, D. and Brunnett, G. (2010). Snake-based segmentation of teeth
from virtual dental casts. Computer-Aided Design and Applications. 7(2), 221233.
Kutsch, V. K. (2011). Caries Detection, Inside Dentistry. AEGIS Communications.
Lai, Y. and Lin, P. (2008). Effective segmentation for dental X-ray images using
texture-based fuzzy inference system. In Advanced Concepts for Intelligent Vision
Systems. Springer, 936947.
Leventon, M. E., Grimson, W. E. L. and Faugeras, O. (2000). Statistical shape
influence in geodesic active contours. In Computer Vision and Pattern Recognition,
2000. Proceedings. IEEE Conference on, vol. 1. IEEE, 316323.
Li, B. N., Chui, C. K., Chang, S. and Ong, S. H. (2011). Integrating spatial
fuzzy clustering with level set methods for automated medical image segmentation.
Computers in Biology and Medicine. 41(1), 110.
Li, S., Fevens, T., Krzyzak, A. and Li, S. (2006). Automatic clinical
image segmentation using pathological modeling, PCA and SVM. Engineering
Applications of Artificial Intelligence. 19(4), 403410.
Lin, P., Huang, P., Huang, P., Hsu, H. and Chen, C. (2014). Teeth segmentation of
dental periapical radiographs based on local singularity analysis. Computer Methods
and Programs in Biomedicine. 113(2), 433 445. ISSN 0169-2607.
Lin P, L., Lai, Y. and Huang, P. (2010). An effective classification and numbering
system for dental bitewing radiographs using teeth region and contour information.
Pattern Recognition. 43(4), 13801392.
Linda, G. S. and George, C. S. (2001). Computer Vision. PrenticeHall.
Lloyd, S. (1982). Least squares quantization in PCM. Information Theory, IEEE
Transactions on. 28(2), 129137.
167
Ma, W.-Y. and Manjunath, B. (1997). Edge flow: a framework of boundary detection
and image segmentation. In Computer Vision and Pattern Recognition, 1997.
Proceedings., 1997 IEEE Computer Society Conference on. IEEE, 744749.
Mahoor, M. H. and Abdel-Mottaleb, M. (2005). Classification and numbering of teeth
in dental bitewing images. Pattern Recognition. 38(4), 577586.
Malladi, R., Sethian, J. A. and Vemuri, B. C. (1995). Shape modeling with front
propagation: A level set approach. Pattern Analysis and Machine Intelligence, IEEE
Transactions on. 17(2), 158175.
Mangin, J.-F., Frouin, V., Bloch, I., Regis, J. and Lopez-Krahe, J. (1995). From 3D
magnetic resonance images to structural representations of the cortex topography
using topology preserving deformations. Journal of Mathematical Imaging and
Vision. 5(4), 297318.
Manousakas, I., Undrill, P., Cameron, G. and Redpath, T. (1998). Split-and-merge
segmentation of magnetic resonance medical images: performance evaluation and
extension to three dimensions. Computers and Biomedical Research. 31(6), 393
412.
Metzler, V. H., Thies, C. and Lehmann, T. M. (2001). Segmentation of medical images
by feature tracing in a self-dual morphological scale-space. In Medical Imaging
2001. International Society for Optics and Photonics, 139150.
Michael, D. J. and Nelson, A. C. (1989). HANDX: a model-based system for
automatic segmentation of bones from digital hand radiographs. Medical Imaging,
IEEE Transactions on. 8(1), 6469.
Mumford, D. and Shah, J. (1989). Optimal approximations by piecewise smooth
functions and associated variational problems. Communications on pure and applied
mathematics. 42(5), 577685.
Nagel, R. (2011). Cure Tooth Decay.
Nassar, D. E. M., Chaudhry, F. U. and Ammar, H. (2004). On Performance Evaluation
of Image Segmentation Algorithms: Success is Not All or None. In In Proceeding,
1st International Computer Engineering Conference (ICENCO 2004), Egypt. Dec.
354359.
Nilsson, B. and Heyden, A. (2003). A fast algorithm for level set-like active contours.
Pattern Recognition Letters. 24(9), 13311337.
Nomir, O. and Abdel-Mottaleb, M. (2005). A system for human identification from
X-ray dental radiographs. Pattern Recognition. 38(8), 12951305.
Nomir, O. and Abdel-Mottaleb, M. (2007). Human identification from dental X-ray
168
images based on the shape and appearance of the teeth. Information Forensics and
Security, IEEE Transactions on. 2(2), 188197.
Nunes, J. C., Guyot, S. and Delechelle, E. (2005). Texture analysis based on local
analysis of the bidimensional empirical mode decomposition. Machine Vision and
applications. 16(3), 177188.
Oliveira, J. and Proena, H. (2011). Caries Detection in Panoramic Dental X-ray
Images. In Computational Vision and Medical Image Processing. (pp. 175190).
Computational Methods in Applied Sciences, vol. 19. Springer Netherlands.
Olsen, G. F., Brilliant, S. S., Primeaux, D. and van Najarian, K. (2009). An
image-processing enabled dental caries detection system. In Complex Medical
Engineering, 2009. CME. ICME International Conference on. IEEE, 18.
Osher, S. and Sethian, J. A. (1988). Fronts propagating with curvature-dependent
speed: algorithms based on Hamilton-Jacobi formulations. Journal of computational
physics. 79(1), 1249.
Otsu, N. (1975). A threshold selection method from gray-level histograms.
Automatica. 11(285-296), 2327.
Paragios, N. and Deriche, R. (2000). Geodesic active contours and level sets
for the detection and tracking of moving objects. Pattern Analysis and Machine
Intelligence, IEEE Transactions on. 22(3), 266280.
Patel, B. C. and Sinha, D. G. (2010). An adaptive K-means clustering algorithm for
breast image segmentation. International Journal of Computer Applications. 10(4),
3538.
Peng, D., Merriman, B., Osher, S., Zhao, H. and Kang, M. (1999). A PDE-based fast
local level set method. Journal of Computational Physics. 155(2), 410438.
Perez, A. and Gonzalez, R. C. (1987). An iterative thresholding algorithm for image
segmentation. IEEE transactions on pattern analysis and machine intelligence. 9(6),
742751.
Pham, D. L., Xu, C. and Prince, J. L. (2000). Current methods in medical image
segmentation 1. Annual review of biomedical engineering. 2(1), 315337.
Phen-Lan, L., Lai, Y.-H. and Huang, P.-W. (2012). Dental biometrics: Human
identification based on teeth and dental works in bitewing radiographs. Pattern
Recognition. 45(3), 934946.
PhongDinh, V. and Le, B. (2008). Dental Radiographs Segmentation Based on Tooth
Anatomy. IEEE, International Conference on Research, Innovation and Vision for
the Future in Computing & Communication Technologies.
169
computer aided lesion detection in dental X-rays using variational level set. Pattern
Recognition. 40(10), 28612873.
Shuo, L., Fevens, T., Krzyzak, A. and Li, S. (2006). An automatic variational level
set segmentation framework for computer aided dental X-rays analysis in clinical
environments. Computerized Medical Imaging and Graphics. 30(2), 6574.
Sirona (2008). SIDEXIS XG. The Dental Company. Retrievable at
http://manuals.sirona.com/en/imaging-systems/software/
sidexis-xg.html.
Smith, M. J. and Docef, A. (1997). A study guide for Digital Image Processing. vol. 1.
Scientific Publishers, Incorporated.
Smith, S. M. and Brady, J. M. (1997). SUSANa new approach to low level image
processing. International journal of computer vision. 23(1), 4578.
Sonis, S. T. (2003). Dental Secrets. Hanley & Belfus.
Sprawls, P. (1995). Physical principles of medical imaging. Medical Physics Pub.
Taheri, S., Ong, S. H. and Chong, V. (2010). Level-set segmentation of brain tumors
using a threshold-based speed function. Image and Vision Computing. 28(1), 2637.
Udupa, J. K. and Samarasekera, S. (1996). Fuzzy connectedness and object definition:
theory, algorithms, and applications in image segmentation. Graphical models and
image processing. 58(3), 246261.
van Bemmel, C. M., Spreeuwers, L. J., Viergever, M. A. and Niessen, W. J. (2003).
Level-set-based artery-vein separation in blood pool agent CE-MR angiograms.
Medical Imaging, IEEE Transactions on. 22(10), 12241234.
Wan, S.-Y. and Higgins, W. E. (2003). Symmetric region growing. Image Processing,
IEEE Transactions on. 12(9), 10071015.
WHO, W. H. O. (2012). Oral health. Retrievable at http://www.who.int/
mediacentre/factsheets/fs318/en/.
Xie, X. (2010). Active contouring based on gradient vector interaction and constrained
level set diffusion. Image Processing, IEEE Transactions on. 19(1), 154164.
Xie, X. and Mirmehdi, M. (2008). MAC: Magnetostatic active contour model. Pattern
Analysis and Machine Intelligence, IEEE Transactions on. 30(4), 632646.
Xingfang, Y., Yumei, H. and Yan, L. (2010). An improved SUSAN corner detection
algorithm based on adaptive threshold. In Signal Processing Systems (ICSPS), 2010
2nd International Conference on, vol. 2. IEEE, V2613.
Xu, C., Yezzi Jr, A. and Prince, J. L. (2000). On the relationship between
171
parametric and geometric active contours. In Signals, Systems and Computers, 2000.
Conference Record of the Thirty-Fourth Asilomar Conference on, vol. 1. IEEE, 483
489.
Yezzi Jr, A., Kichenassamy, S., Kumar, A., Olver, P. and Tannenbaum, A. (1997).
A geometric snake model for segmentation of medical imagery. Medical Imaging,
IEEE Transactions on. 16(2), 199209.
Yoon, O.-K., Kwak, D.-M., Kim, D.-W. and Park, K.-H. (1999). MR brain image
segmentation using fuzzy clustering. In Fuzzy Systems Conference Proceedings,
1999. FUZZ-IEEE99. 1999 IEEE International, vol. 2. IEEE, 853857.
Zhang, K., Zhang, L., Song, H. and Zhou, W. (2010). Active contours with selective
local or global segmentation: a new formulation and level set method. Image and
Vision computing. 28(4), 668676.
Zhou, J. and Abdel-Mottaleb, M. (2005). A content-based system for human
identification based on bitewing dental X-ray images. Pattern Recognition. 38(11),
21322142.
Zhu, F. and Tian, J. (2003). Modified fast marching and level set method for medical
image segmentation. journal of X-Ray Science and Technology. 11(4), 193204.