DENTAL information from texture, shape, contours, etc is used

 DENTAL IMAGE PROCESSING TECHNIQUES – A LITERATURE REVIEW Rameswari Poornima Janardanan Abstract —Early detection and diagnosis of diseases are facilitated by medical imageprocessing methods. It is a strong tool aiding medical research and relatedclinical practices.

An add on approach survey is done in this paper in the areaof medical image analysis for diagnosis of diseases in oral radiology usingdental x-rays in dentistry. The interpretations of medical images rely hugelyon human involvement and the human perception of the details present init.  The interpretation of the delicatefine details in various contrast situations present in a medical image isindeed a cumbersome task to assess. Typical radiographs obtained from a regularradiograph acquisition device may be of poor or average quality inrepresentation. Various standardized scientific tools have been designed byresearchers, scholars and software developers to address this type ofshortcomings in a medical radiograph. These are targeted to minimize thepossible human error in predicting the right diagnosis and treatment solely onthe basis the human visual perception.

Feature extraction on focused area forthe information required, on an extracted tooth area in a digital dentalradiograph is highlighted in this feature of utmost support to a dentist for pre-diagnosisat an early stage.1.  IntroductionImage processing includes several methods like enhancement,segmentation, region of interest detection, filtering methods, thresholdingtechnique and morphological operations. The information from texture, shape,contours, etc is used in the classical image segmentation. Edge detection isused to find boundaries of objects inside an image. Image enhancementtechniques are used to restore the original image.

              Medical imagingtechnology has revolutionized the health care over the past three decades, aiding doctors todiagnose and improve patient outcomes. A fight against cancer is fought effectively using medicalimaging in its prevention, diagnosis and, treatment. An important advantage of digital dental radiography isits ability to process the image data, so that the information content of theimage is more accessible to the human visual system. Dentalprofessionals today are increasingly using digital dental x-rays for betterdetection, diagnosis, treatment and monitoring of oral conditions and diseases.Traditional x-ray films are replaced by the digital electronic sensors.

Thesesensors can produce enhanced computer images of intra oral structures andconditions.               The aim of this systematicreview is to give an overview towards current dental image processing methodsbecause of their potential importance in the dental and forensic fields. Therefore,this paper is sectioned and sub sectioned as section 3.2  reviews the methodology a review of image enhancementmethods used on dental X-ray images. Section 3.

3reviews various techniques usedfor image segmentation and feature extraction on dental radiographs. Thissection also highlights the works done in forensic odontology using imagesegmentation. Section 3 concludes the present review. 1.

1 Why it is important to do this review The relevance of this review is grounded on the need torecommend a method for dental age estimation and human identification with thefollowing characteristics: simple, fast, non-invasive, non-expensive,reproducible and over all, accurate, that can be systematically used indifferent academic and forensic scenarios. This efficiently assists in identifying deceased individuals oridentify human profiles in any doubtful situations 2.   Methodology There had been many trials todevelop automated computer vision based systems to facilitate forensic odonatologicalapplications. These systems comprises of variety of image processingtechniques.  The basic algorithms andmethods used in dental x-ray processing are image enhancement, image segmentation,edge detection with feature extraction and neural networks based classification.The inclusion criteria were the studies with imageprocessing techniques with application to dental clinical and forensicapplications.

The eligibility criteria are as shown in the Table 1.Table 1. Summary of included imageprocessing methods used on dental radiographs and its purpose. Authors Image processing methods used Purpose of the study                                       Studies which had similar methods and often usedwere excluded.  Table 2. The list of data that was extractedfrom the reviewed full texts. Data extracted from full text items First author Year Title Segmentation method Segmentation software Clinical application  2.1 Study identification andselectionThe information was searched throughthe data base available through the Saudi digital library accessed through thee-library facilitated by Riyadh Colleges of Dentistry and Pharmacy.

Directoryof open access journals(DOAJ), Medline/PubMed (NLM), ProQuest, Collection, (Webof science), Science Direct Journals(Elsevier), Wiley(Cross Ref),Wiley Onlinelibrary, google scholar were accessed to assimilate information  this review. Reviews, articles, reports andoriginal papers published in peer journals, books, conference proceedings forgrey literature were all considered. English language publications from anysetting and recent time frame from 2010 till date, were considered eligible.The search keywords used were dentalimage processing, image segmentation on dental radiographs, humanidentification from dental x-rays, dental age estimation methods 2.

2Dataextraction and managementThe collected information was organized in anexcel spread- sheet as follow: Author, year,country, number of participants, image processing algorithm used,and itsapplications.  2.3 Eligibility criteria forconsidering in this reviewThe scope of this review was notlimited to general dental image processing methods, but a brief description ofits clinical and forensic applications were reviewed  2.4 Assessment of risk of bias in includedstudiesTo avoid bias in this systematic review, and to avoid falsepositive conclusions or falsenegative it was necessary to analyze the possibility of author bias. This owedto the participation of the same authors in repeated publications. To thisend, the resultswere analyzed comparingindividual papers, and then grouping them per author.  Fig. 1.

Flowchart of the study selection this review 3. Review A review of imageenhancement methods used on dental X-ray imagesThe physicalprocess of digital radiography is quite similar to traditional dental x-raysthat use films. A digital electronic senor is used to capture images of theoral cavity and its structures. This is connected to a computer so that oncethe x-ray is taken, the image can be projected on a screen for the dentist toview. Dental imagesare typically classified into periapical, panoramic, and bite wing dentalimages, as shown in figure(1) and figure (2) and figure (3).Bitewing images are most preferred for dental processing.A digital radiograph has theadvantage of immediate image preview and availability, and eliminates the costof film processing steps.

It provides the ability to apply special imageprocessing techniques that enhance the overall display quality of the image andextract only the regions of interests.  3.1 Reviews various techniques usedfor image segmentation and feature extraction on dental radiographsEyadHaj presented an over view about an automated dental identification system forhuman identification1. This dental identification system can beused by both law enforcement and security agencies in both forensic andbiometric identification. The various techniques for dental segmentation of X-ray images to address the problem of identifying each individual tooth and howthe contours of each tooth are extracted is presented. Their technique was notable to properly segment an X- ray by a single segmentation technique and itvaried from image to image.

Areview on dental biometric systems and technology with further applications inforensic science was done. In  the paper by S.Kiattisin,2008 the authorspresent a match of X-ray teeth films using image processing based on specialfeatures of teeth. This method helps the dental doctors to match simply a pairof teeth using the special features of the teeth films. Teeth’s pictures arescanned and adjusted by a scanner and a computer, respectively, as well as thenthey are converted into binary code and decoded to the direction code (chaincode). Chain code is a method for decoding a direction code from the binaryimages based on the special features of teeth.

The chain code of each pictureis compared with the statistical chain code. Therefore, the percentage of thesame chain code is approximately 90% (i.e. matching same patterns) for thecomparison of one root to one root (7 times) and two roots to two roots (7times) while the percentage of the same chain code is reduced at relativelybelow 50% (i.e. matching different patterns) for comparison of one root to tworoots (2 times). The percentage of the same chain code is approximately 90%(i.e.

matching same patterns) for the comparison of one root to one root (7times) and two roots to two roots (7 times) while the percentage of the samechain code is reduced at relatively below 50% (i.e. matching differentpatterns) for comparison of one root to two roots (2 times).In(Maja Omanovic,2008)  the  sum of squared differences(SSD) cost functionshows the degree of similarity or overlapping between two radiographs degree ofsimilarity/overlap between two radiographs .This method was tested on adatabase of 571 radiographs belonging to 41 distinct individuals. Figure 4shows an overview of this process. A total of 150 identification scenarios weretaken then each single ROI was identified/extracted for comparing andmatching with the dental x- ray images.   Figure 4 ? Illustration of the identification test runn& topthree radiographs in the database ranked by the associated cost.

(MajaOmanovic,2008) Theauthors proposed a computer-aided framework for matching of dental radiographsbased on a sum of squared differences cost criterion. In their framework, theoperator would define the ROI by roughly circling the tooth of interest on agiven post-mortem radiograph. Hence, even untrained staff able to participatein the identification efforts by roughly circles the tooth area. The systemitself then matches the selected region to radiographs found in the ante mortemdatabase. For all possible shifts, the best brightness and contrast adjustmentand rotation were computed, and the parameters that yielded.The lowest cost arerecorded along with the associated cost (match score). The radiographs in thedatabase were then ranked according to the cost, with the lowest costindicating the best match.

This work was not tested on multiple ROI’s as wellas on different dental images.(S.Lailee,2005)Investigated the fundamental problems in image segmentation using traditionalsegmentation techniques and proposed an improved technique for segmentingimages captured under natural environment. Due to non-uniform illumination itis difficult to produce a significant threshold value along with lack ofdifference in reflection. Since different illumination may produce differentcolor intensity of the object surface and thus lead to inaccurate segmentedimages. The widely used traditional method for thresholding is ostu and fuzzyc-means respectively. In this method, the authors have added a step extra afterthresholding with ostu method by converting the gray scale image into binary& then integrating the  modifiedthreshold algorithm with an inversion technique.

The results were analyzedbased on rand index function. By this the authors have concluded that theimages after ostu method and thresholding which were not able to get separateand provide the required information are now being able to separate theinterest area & background easily. The ability of this technique thereforehas the potential to classify the poor images with inconsistent illumination condition.Dental biometrics can be used in forensic sciencefor human identification.

It utilizesdental radiographs. This radiograph providesinformation related to tooth shape,teeth contour and relative position of neighboring teeth,also gives shapes of dental worklike crowns, filling  & bridges etc. Dentalbiometrics requires ante mortem (AM) and post mortem (PM) radiographs forfinding unidentified subject. Dental biometrics having three stages:Pre-processing and segmentation of radiographs, contour extraction or dentalwork extraction, atlas registration and matching.

Segmentation can be done byvarious methods. Contour or shape of teeth and dental work can be extracted.Method or code was developed by the authors to locate teeth this is known asdental atlas registration. Numbering to teeth from left to right of jaw andalso  differentiation between upper jawand lower jaw was done, which help inthe matching stage (S.Jadhav,2012) (Omaima Nomir ,2005) presents a system inwhich, given a dental image of a post-mortem (PM), the proposed systemretrieves the best matches from an ante mortem (AM) database. The systemautomatically segments dental X-ray images into individual teeth and extractsthe contour of each tooth.

Features are extracted  from each tooth and are used for retrieval.During retrieval, the AM radiographs that have signatures closer to the PM arefound and presented to the user. Matching scores are generated based on thedistance between the signature vectors of AM and PM teeth.Figure5. Block diagram of segmentation algorithm.

(Omaima Nomir ,2005) Theyintroduced iterative and adaptive thresholding. Thereafter horizontal andvertical integral projection is used for separating the jaws as well asindividual tooth. The block diagram of segmentation algorithm is as shown inFigure 5.

This technique was not successful in matching images due to poorquality of images and shape of teeth could have changed with time as PM imageswere taken after a long time AM images were captured. (Said,E.H,2006)in his paper designed an approach based on mathematical morphological segmentation. Greyscale contrast stretchingtransformation is performed for an enhanced teeth segmentation performance.

Itpresented a technique with a low failure rate on comparison to otherapproaches. Figure 6 Mainstages of the algorithm(Said,E.H,2006)Figure .

7Grayscale line profiles of the input image, the upper horizontal line profileillustrates the bones between the teeth, the lower horizontal line profileshows the gap between the teeth, while the vertical line profile illustrates thegap valley. (Said,E.H,2006) (Chen and Jain2005) Thetooth contour is the feature extracted as they remain invariant over time incomparison to other feature of the teeth.

Radiograph segmentation and contourextraction are done in the feature extraction stage. Based on edge detectioncontour extraction is approached. Figure 9 .Theprocessing flow diagram(Chen and Jain2005) andthe results of teeth alignment and dental work alignment with the parametersused in teeth alignment. (a) Query DW. (b) Genuine DW. (c) Imposter DW.

(d) Thecontours of the DW in (b) and the DW in (a) being affine transformed with theteeth alignment parameters between (a) and (b). (e) The contours of the DW in(c) and the DW in (a) being affine transformed with the teeth alignmentparameters between (a) and (c). (Chen and Jain2005) (Hofer, Marana,2007) proposed a method to perform human identification based on dental workinformation. The algorithm involves 3 steps namely Segmentation, Featureextraction, Creation of a dental code and matching.The dental code includes theinformation about the upper and lower jaw position, sizes of the dental workand the distance between two neighboring dental works. Maxillary and mandibularteeth border is detected and then the intensity sum of all horizontal rows inthe strip is calculated. The dental work is identified by the highestintensity.

 Figure10.  Cut stripe (region 1) and sum of intensities;right valley represents lower intensity which indicates that the DW belongs tothe maxilla teeth (dental code = “U”). (Hofer, Marana, 2007)The firstvalley on the right and left site of the maximum intensity point is detected bythe algorithm.

The lower intensity valley represents the mandibular andmaxillary teeth border.3.  Results and conclusionThe major researchers make use of thresholding and morphologicaloperation for feature extraction and segmentation.

However, in the existingsoftware’s used by doctors the option of adaptive or global threshold is notavailable. Hence, the benefits of these methods are not directly available.Much of the work have been done for human identification, but very fewresearchers have applied and realized the methods for diagnosis purpose. Forthe diagnosis of intra oral diseases specifically the region of interestselection, impacted 3rd molar using x ray rendering of 3-D images and other relatedproblems of gums and idiopathic resorption is still a missing feature in mostof the software’s. Interactive portions of X-ray selected for furtherprocessing specifically for the purpose ofdiagnosisis the need of the hour as it would helpboth doctor and patient to understand the problem and depth of disease. Nosoftware is using AI tools such as neural networking, fuzzy c-means, etc. for the better understanding and diagnosis purpose.

Researchers up till now have beenfound concentrating on image enhancement or segmentation forextracting features for forensic sciences. No much research has effectivelycontributed for the diagnostic methods. Automated or semi-automated diagnosis of aforesaid objectives would be quiet useful for doctor as well aspatient. Image processing & enhancementfunctions are rarelyincorporated in commercial software for direct digitalimaging in dental radiology.

Until now, comparison of software was limitedby arbitrary namingused in each system.Standardized terminology and increased functionality of image processing should be offered to the dental profession.This systematic review summarizes and compares the results of some of the most used methods for dental ageestimation in adults, performing a qualitative and quantitative analysis.—————————————. Age estimation in adults is a challenge in all forensiccontexts, especially in cases that require the use of non-invasive methods.————————-In the light of the evidence one could ———————.

————————REFERENCES 1   E. H, said,G. Fahmy, D. nassar, H.

Amar,”Dental X-ray image segmentation” Biometric Technology for HumanIdentification, Proceedings of the SPIE, Vol. 5404, pp. 409417,2004.2  E. H. Said, D. E. M.

Nassar, G. Fahmy,H. H. Ammar. “Teeth segmentation in digitizeddental X-ray films using mathematical morphology,” IEEE Transactions on informationforensic and security, vol.

1, Issue. 2, pp. 178189, June. 2006.3  Hofer, M. and Marana, A.N.

, 2007, October. Dental biometrics:human identification based on dental work information. In ComputerGraphics and Image Processing, 2007. SIBGRAPI 2007. XX Brazilian Symposium on (pp.281-286). IEEE.

4   Maja Omanovic, Jeff J. Orchard “ExhaustiveMatching of Dental X-rays for Human ForensicIdentification “Journal of the Canadian Society of Forensic Science, 2008 16 5   S.Jadhav, R. Shriram “Dental biometrics used inforensic science” IJERS/Vol.III/ Issue I/January-March, 2012/26-29 6   Omaima Nomir, M.

A.Mottaleb “A system for humanidentification from X-ray dental radiographs” Pattern Recognition 38 (2005)1295 – 1305.7   S. Dighe, R. Shriram, “Pre-processing,Segmentation and Matching of Dental Radiographs used in Dental Biometrics”,International Journal of Science and Applied Information Technology,Volume 1, No.2,pp 52- 56, May – June 20128   S. Kiattisin, A. Leelasantitham, K.

Chamnongthai , K. Higuchi, ” A Match of X-ray Teeth Films Using Image processingBased on Special Features of Teeth” , SICE Annual Conference 2008, pp 35-399   ?. Oprea, C. Marinescu , I. Li??, M. Jurianu,D.

A. Vi?an, I. B. Cioc, ” Image Processing Techniques used for Dental Xray Image Analysis” , Electronics Technology, ISSE2008, pp 125-12910  S.

A.Ahmed, M.N.

Taib, N.E.A.Khalid, R.Ahmad,H.Taib “Performance of compound enhancement algorithms on dental radiographimages” WASET-2011 1311                      Said, E.H., Nassar, D.

E.M., Fahmy, G. and Ammar, H.H., 2006.Teeth segmentation in digitized dental X-ray films using mathematicalmorphology. IEEE transactions on information forensics and security, 1(2),pp.

178-18912  Stefan Michel, Saskia M.Koller, Markus Ruh,Adrian Schwaninger, “Do “Image Enhancement” Functions Really Enhance X-ray Image Interpretation?” Cognitive ScienceJournal Archieve 200713  Shafer’s Tb. “Textbook of Oral Pathology” sixthedition, 2006.14  Sharifah Lailee, Nursuriati Jamil “Segmentationof Natural Images Using an Improved Thresholding-based Technique” IRIS 2012, Elsevier Procedia engg. Conference pp938-944.15  T. N. Cornsweet, Visual Perception, AcademicPress, New York, 1970.

16  TM Lehmann, E.Troeltsch, K Spitzer” Image processing and enhancement providedby commercial dental software programs: A Technical Report”Dentomaxillofacial Radiology (2002) 31, 264-272.17  White & Pharoah “Oral Radiology- Principlesand Interpretation”, Fifth Edition (2005), selected illustration by Dr. DonaldO’Connor.

ISBN 0-323-02001,published by MOSBY ( An affiliate of Elsevier)