Tissue Classification using Cluster Features for Lesion Detection in Digital Cervigrams
|
|
- Harold Johnston
- 5 years ago
- Views:
Transcription
1 Tssue Classfcaton usng Cluster Features for Leson Detecton n Dgtal Cervgrams Xaole Huang 1, We Wang 1, Zhyun Xue 2, Sameer Antan 2, L. Rodney Long 2, Jose Jeronmo 3 1 Department of Computer Scence and Engneerng, Lehgh Unversty, PA, USA 2 Communcatons Engneerng Branch, Natonal Lbrary of Medcne, MD, USA 3 Dvson of Cancer Epdemology and Genetcs, Natonal Cancer Insttute, MD, USA ABSTRACT In ths paper, we propose a new method for automated detecton and segmentaton of dfferent tssue types n dgtzed uterne cervx mages usng mean-shft clusterng and support vector machnes (SVM) classfcaton on cluster features. We specfcally target the segmentaton of precancerous lesons n a NCI/NLM archve of 60,000 cervgrams. Due to large varatons n mage appearance n the archve, color and texture features of a tssue type n one mage often overlap wth that of a dfferent tssue type n another mage. Ths makes relable tssue segmentaton n a large number of mages a very challengng problem. In ths paper, we propose the use of powerful machne learnng technques such as Support Vector Machnes (SVM) to learn, from a database wth ground truth annotatons, crtcal vsual sgns that correlate wth mportant tssue types and to use the learned classfer for tssue segmentaton n unseen mages. In our experments, SVM performs better than un-supervsed methods such as Gaussan Mxture clusterng, but t does not scale very well to large tranng sets and does not always guarantee mproved performance gven more tranng data. To address ths problem, we combne SVM and clusterng so that the features we extracted for classfcaton are features of clusters returned by the mean-shft clusterng algorthm. Compared to classfcaton usng ndvdual pxel features, classfcaton by cluster features greatly reduces the dmensonalty of the problem, thus t s more effcent whle producng results wth comparable accuracy. Keywords: Image segmentaton, color space, support vector machnes, clusterng, features, tssue classfcaton, leson detecton, dgtal cervgrams, cervcal cancer 1. INTRODUCTION To make mages searchable by content n large medcal archves, t s very mportant to relably segment and label dfferent tssue regons, especally bomarker regons. We consder the automated segmentaton problem n a very large archve of 60,000 dgtzed uterne cervx mages, created by the Natonal Lbrary of Medcne (NLM) and the Natonal Cancer Insttute (NCI). These mages are optcal cervgram mages acqured by Cervcography usng specally desgned cameras for vsual screenng of the cervx, and they were collected from the NCI Guanacaste proect for the study of vsual features correlated to the development of precancerous lesons. The most mportant observaton n a cervgram mage s the Acetowhte (AW) regon, whch s caused by whtenng of potentally malgnant regons of the cervx epthelum, followng applcaton of acetc acd to the cervx surface. The cervgram mages n the archve have large varatons n ther appearance due to llumnaton varatons, artfacts n mage acquston, and ntrnsc dfferences n mage content. Exstng methods for cervgram mage analyss [1,2,3] consst of sequental steps of mage processng such as pre-processng to remove specular reflecton, segmentng
2 Fgure 1. AW vs. Non-AW Cervx color sample dstrbutons n L*a*b* space. (a) samples from one mage. (b) samples from two mages. (c) samples from three mages. (d) samples from sx mages. (red) AW color samples. (blue) Cervx color samples. cervx boundary, detecton of OS, detecton of columnar epthelum, thresholdng, and mosac texture analyss. These methods acheved good results on cervx area detecton and on fllng n specular regons, but the performance on segmentaton of mportant tssue regons such as acetowhte and columnar epthelum needs mprovement. Furthermore, due to large varatons n mage appearance, color and texture features of a tssue type n one mage often overlap wth that of a dfferent tssue type n another mage. Ths makes relable segmentaton n a large number of mages a very challengng problem. A popular color-based tssue segmentaton method s to apply clusterng technques such as K-means [3], Gaussan Mxture Models [2] and Mean-shft [6], to drectly model the posteror probabltes p(c e), where e represents an evdence vector that descrbes mage features (e.g. pxel color), and c = 1,...,C s one of the C tssue classes. One challenge facng clusterng methods n large-scale segmentaton s that color dstrbuton of one tssue class from many mages can have many modes and overlap sgnfcantly wth color dstrbutons of other tssue classes. Fgure 1 demonstrates ths problem by dsplayng Acetowhte (AW) and cervx (non-aw) color samples from 1, 2, 3, and 6 mages. Note that, as the number of mages ncreases, the AW and cervx samples ncreasngly overlap wth each other. It s therefore dffcult to predct the class of a test color sample wthout a hgh probablty of error gven assumptons about the color dstrbutons of the tssue classes. In addton, not every tssue type s always present n every mage, hence there wll lack relable ways to automatcally set clusterng parameters such as the number of clusters n K-means and GMM, and the sze of the bandwdth n Mean shft. In ths paper, we propose a database-guded segmentaton paradgm n whch we apply machne learnng technques, such as support vector machnes (SVM) to learn, from a database wth ground truth annotatons provded by experts, crtcal vsual sgns that correlate wth mportant tssue types and to use the learned classfer for tssue segmentaton n unseen mages. The support vector machnes (SVM) classfer [4, 5] has been successfully appled to detectng Mcrocalcfcatons n Mammograms [8] and varous other medcal classfcaton problems. In ths paper we use SVM to perform color-based tssue classfcaton n order to segment dfferent tssue regons, especally to segment the bomarker acetowhte (AW) regon from the rest of the cervx. The segmentaton performance s optmzed wth respect to the feature color space and granularty. We evaluated color spaces ncludng RGB, HSV, and L*a*b*. On dfferent granularty of the features, we tran AW and other tssue classfers, frst usng ndvdual pxel sample colors and then usng cluster features returned by the Mean Shft based clusterng algorthm [6]. Cluster features greatly reduce the dmensonalty of tranng so that SVM s scalable to larger tranng sets, whle producng results wth comparable accuracy. Gven a novel test mage, the Mean Shft clusterng algorthm parttons the mage nto clusters of smlar color and/or texture, and the traned SVM classfer (on cluster features of tranng data) s appled to classfyng clusters n the test mage. Ths ground-truth database guded segmentaton method s flexble n terms of the number of tssue classes. Thus we can perform ether two-label (e.g. AW vs. Non-AW cervx), or mult-label (e.g. AW, CE, SE, other) classfcaton.
3 2. METHODOLOGY For tranng and testng purposes, we have access to ground truth boundary markngs on 939 cervgram mages from the NCI/NLM archve. The ground truth markngs are collected usng a web-based Boundary Markng Tool developed by NLM and NCI [7]. There were 20 expert evaluators who used the tool to manually outlne AW and Cervx boundares n the 939 cervgrams. Some cervgrams have boundares annotated by multple experts; n ths case, we randomly choose one as the ground truth although mechansms for combnng multple-expert annotatons are avalable [9]. Fg. 1 shows some example cervgram mages wth expert annotatons. Fgure 2. Example cervgram mages wth boundary markngs by experts. the yellow outlnes cervx boundary. The blue outlnes AW regons and Fgure 3. Dstrbuton of pxel color samples n RGB (left) and L*a*b* (rght) color spaces Color Features Color and texture are two most promnent features for tssue classfcaton n dgtal cervgrams. In ths paper, we nvestgate the color features n dfferent color spaces and at dfferent granulartes. Choces of color spaces nclude the RGB, HSV, CIE L*a*b*, and others. Our experments show that lumnance (or ntensty) s an mportant feature n dscrmnatng between tssue classes, thus HSV and L*a*b* are preferred color spaces. Because of the quantzaton dscontnuty n the hue dmenson (e.g. 255 and 0 hue values are both very close to red) n the HSV color space however, for segmentaton we choose to use pxel colors n the CIE L*a*b* space. Dstrbuton of color samples n the L*a*b* space s also better for clusterng and classfcaton [6], as shown n Fg. 2.
4 Fgure 4. AW segmentaton examples. Test mage (top row), segmented AW regons (bottom row) SamplesPerImage * # of Images Tranng Tme Memory Usage 200*10 67 sec 16MB 200*20 18 mns 53MB 200*30 1 hour 34 mns 103MB 200*55 20 hours 20mns 362MB Table 1. SVM tranng tme and memory usage gven tranng sets of dfferent szes Pxel sample features vs. Cluster features Pxel color features for tranng and classfcaton A two-class pxel-wse color classfer s traned by selectng Acetowhte (AW) pxel colors n the marked ground truth AW areas as postve samples, and selectng Cervx (non-aw) pxel colors as negatve samples. 55 mages are used for tranng and 120 mages for testng. A confdence-rated SVM classfer wth a lnear kernel [4] s traned to dfferentate AW from non-aw pxel colors. Gven a test mage, the classfer s appled drectly to mage pxels, and all pxels havng the confdence rate above zero are consdered as part of the AW regon. Usng false postve fracton (FPF) and false negatve fracton (FNF) for quanttatve evaluaton, the pxel-wse classfer acheved an average of 23% FPF and 9% FNF. Some examples of AW segmentaton usng the pxel-wse classfer are shown n Fg Mean-shft clusterng and usng cluster centers for tranng and classfcaton The SVM learnng on pxel color features produces promsng AW classfcaton results as shown above. However, before SVM learners can potentally become a soluton for tssue (especally bomarker tssue) segmentaton n large medcal mage archves, we need to address the scalng problem. We record n Table 1 the processng tme and memory usage gven tranng sets of dfferent szes. One can see that, as the number of mages and/or the number of pxels from each mage for tranng ncrease, the SVM tranng tme and memory usage explode exponentally. To solve ths problem, we experment wth cluster or regon features nstead of ndvdual pxel features. For each tranng mage, we frst apply mean-shft clusterng based on L*a*b* color feature and spatal proxmty [6] to group pxels n the mage nto clusters. The label of each cluster (AW or Cervx non-aw) s assgned automatcally based on
5 Fgure 5. Labeled prmary cluster centers of a tranng mage. Feature SamplesP erimage * # of Images Preprocessng (clusterng) tme SVM Tranng Tme Memory use Pxel color 200* hours 20mns 362MB Cluster mean color 30*55 36 mns < 1 mn 12MB Table 2. Performance comparson: cluster centers as tranng data vs. pxel colors as tranng data expert boundary markngs. The cluster center, whch refers to the mean color of all pxels n the cluster, s then taken as the tranng sample. Fg. 5 shows labeled cluster centers for dfferent tssues n one tranng mage. Usng approxmately 30 cluster centers (from 15 largest AW clusters and 15 largest Cervx clusters) per mage, the tranng tme and memory cost are sgnfcantly reduced (Table 2). Gven a test mage, t s frst parttoned nto clusters usng mean shft. Then the SVM classfer learned on cluster centers s appled to classfyng each cluster n the test mage (usng the cluster center feature) to ether AW or Cervx. In our experments, the segmentaton accuracy on AW usng cluster-based classfcaton s comparable to that usng pxel-based classfcaton, whle cluster classfcaton s much more effcent and requres less memory (see Tables 1 and 2) Multple-label Classfcaton Instead of two classes (AW vs. cervx), we perform mult-label classfcaton to segment smultaneously several sgnfcant tssue regons n cervgrams ncludng the Acetowhte (AW), Columnar Epthelum (CE) and Squamous Epthelum (SE). The mult-label classfer s learned based on the one-aganst-one approach [10]. Frst, k*(k-1)/2 classfers are traned, each usng data from two dfferent classes. In a votng strategy, each bnary classfcaton by a two-class classfer s consdered to be a votng where votes can be cast for all data ponts. In the end, a data pont s
6 Fgure 6. Test result for multple tssues by usng RBF kernels of SVM. Frst s the orgnal mage; second s AW; thrd s CE. Fgure 7. Usng dfferent kernels of SVM for two-label (AW vs. cervx) classfcaton. Orgnal mage (upper left), AW by lnear kernel (upper rght), AW by polynomal wth d=3 (lower left), RBF kernel (lower rght). labeled to be n the class wth maxmum number of votes. One example output of segmented AW and CE regons s shown n Fg Kernel functon selecton of SVM We consder several kernel functon selectons for the support vector machnes classfers. Lnear: K x, x ) = x T x (. Polynomal: K( x, x ) = ( γx x + r), γ > 0. 2 Radal bass functon (RBF): K ( x, x ) = exp( γ x x ), γ > 0. Sgmod: K(, x ) = tanh( x x + r) T x γ. T
7 Our emprcal studes show that SVM classfcaton s senstve to kernel selecton, especally when we use cluster-center features because the number of samples for tranng s fewer. Fg. 7 shows an example demonstratng result dfferences by dfferent kernels. Usng cluster center features, our experments show that the RBF kernel outperforms others n two-label classfcaton, whle the lnear kernel s the best n mult-label classfcaton. Usng ndvdual pxel color features, the segmentaton accuracy on most tssues, such as AW, CE, and SE, s comparable by dfferent kernels on most test mages. 3. CONCLUSIONS AND DISCUSSIONS We ntroduce a database guded dscrmnatve approach to segmentng tssue, especally bomarker acetowhte tssue, regons n dgtzed uterne cervx mages. Tranng a support vector machne (SVM) classfer usng cluster center features gves us better effcency than usng ndvdual pxel features due to the reduced dmensonalty whle producng comparable accuracy. The method can be extended to segmentng other sgnfcant tssue regons n cervgrams ncludng the Columnar Epthelum (CE) and Squamous Epthelum (SE) usng multple label classfcaton. Comparng dfferent kernel functons for the support vector machnes classfer, we fnd that, wth cluster features, the lnear kernel s more sutable n mult-label classfcaton whle the Radal Bass Functons kernel s better for two-label classfcaton. ACKNOWLEDGMENTS We would lke to thank the Communcatons Engneerng Branch, Natonal Lbrary of Medcne, and the Natonal Cancer Insttute for provdng the data and support of ths work. REFERENCES [1] Zmmerman G. and Greenspan, H., Automatc detecton of specular reflectons n uterne cervx mages, Proc. of SPIE Medcal Imagng, volume 6144, pages (2006) [2] Gordon S., Zmmerman, G., Long, R., Antan, S., Jeronmo J. and Greenspan, H., Content Analyss of Uterne Cervx Images: Intal steps towards content based ndexng and retreval of cervgrams, Proc. of SPIE medcal magng, volume 6144, pages (2006) [3] Tulpule, B., Hernes, D., Srnvasan, Y., Yang, S., Mtra, S., Srraa, Y., Nutter, B., Phllps, B., Long, L.R. and Ferrs, D., A probablstc approach to segmentaton and classfcaton of neoplasa n uterne cervx mages usng color and geometrc features, Proc. of SPIE Medcal Imagng, Volume 5747, pages (2005) [4] Joachms, T., Makng large-scale SVM Learnng Practcal, Advances n Kernel Methods - Support Vector Learnng, B. Schölkopf and C. Burges and A. Smola (ed.), MIT Press (1999) [5] Vapnk, V., Statstcal Learnng Theory, Wley-Interscence (1998) [6] Comancu, D. and Meer, P., Mean shft: A robust approach toward feature space analyss, IEEE Transacton on Pattern Analyss and Machne Intellgence, Vol. 24, No. 5 (2002) [7] Jeronmo, J., Long, L., Neve, L., Bopf, M., Antan, S. and Schffman, M., Dgtal tools for collectng data from cervgrams for research and tranng n colposcopy, J. of Lower Gental Tract Dsease 10(1) (2006) [8] El-Naqa, I., Yang, Y., Wernck, M. N., Galatsanos, N. P. and Nshkawa, R. M., A Support Vector Machne Approach for Detecton of Mcrocalcfcatons," IEEE Trans. on Medcal Imagng, 21(12): (2002)
8 [9] Warfeld, S. K., Zou, K. H. and Wells, W. M., Smultaneous Truth and Performance Level Estmaton (STAPLE): An Algorthm for the Valdaton of Image Segmentaton," IEEE Trans. on Medcal Imagng, 23(7): (2004) [10] Knerr, S., Personnaz, L. and Dreyfus, G., Sngle-layer learnng revsted: a stepwse procedure for buldng and tranng a neural network," In J. Fogelman, edtor, Neu-rocomputng: Algorthms, Archtectures and Applcatons. Sprnger-Verlag (1990)
FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur
FEATURE EXTRACTION Dr. K.Vjayarekha Assocate Dean School of Electrcal and Electroncs Engneerng SASTRA Unversty, Thanjavur613 41 Jont Intatve of IITs and IISc Funded by MHRD Page 1 of 8 Table of Contents
More informationLearning the Kernel Parameters in Kernel Minimum Distance Classifier
Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department
More informationMachine Learning 9. week
Machne Learnng 9. week Mappng Concept Radal Bass Functons (RBF) RBF Networks 1 Mappng It s probably the best scenaro for the classfcaton of two dataset s to separate them lnearly. As you see n the below
More informationContent Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers
IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth
More informationEdge Detection in Noisy Images Using the Support Vector Machines
Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona
More informationClassifier Selection Based on Data Complexity Measures *
Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.
More informationOutline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1
4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:
More informationClassification / Regression Support Vector Machines
Classfcaton / Regresson Support Vector Machnes Jeff Howbert Introducton to Machne Learnng Wnter 04 Topcs SVM classfers for lnearly separable classes SVM classfers for non-lnearly separable classes SVM
More informationTerm Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task
Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto
More informationA Binarization Algorithm specialized on Document Images and Photos
A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a
More informationBOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET
1 BOOSTING CLASSIFICATION ACCURACY WITH SAMPLES CHOSEN FROM A VALIDATION SET TZU-CHENG CHUANG School of Electrcal and Computer Engneerng, Purdue Unversty, West Lafayette, Indana 47907 SAUL B. GELFAND School
More informationEYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS
P.G. Demdov Yaroslavl State Unversty Anatoly Ntn, Vladmr Khryashchev, Olga Stepanova, Igor Kostern EYE CENTER LOCALIZATION ON A FACIAL IMAGE BASED ON MULTI-BLOCK LOCAL BINARY PATTERNS Yaroslavl, 2015 Eye
More informationThe Research of Support Vector Machine in Agricultural Data Classification
The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou
More informationCHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION
48 CHAPTER 3 SEQUENTIAL MINIMAL OPTIMIZATION TRAINED SUPPORT VECTOR CLASSIFIER FOR CANCER PREDICTION 3.1 INTRODUCTION The raw mcroarray data s bascally an mage wth dfferent colors ndcatng hybrdzaton (Xue
More informationA Fast Content-Based Multimedia Retrieval Technique Using Compressed Data
A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,
More informationSRBIR: Semantic Region Based Image Retrieval by Extracting the Dominant Region and Semantic Learning
Journal of Computer Scence 7 (3): 400-408, 2011 ISSN 1549-3636 2011 Scence Publcatons SRBIR: Semantc Regon Based Image Retreval by Extractng the Domnant Regon and Semantc Learnng 1 I. Felc Raam and 2 S.
More informationSmoothing Spline ANOVA for variable screening
Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory
More informationSupport Vector Machines
Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned
More informationA PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION
1 THE PUBLISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Seres A, OF THE ROMANIAN ACADEMY Volume 4, Number 2/2003, pp.000-000 A PATTERN RECOGNITION APPROACH TO IMAGE SEGMENTATION Tudor BARBU Insttute
More informationMachine Learning. Topic 6: Clustering
Machne Learnng Topc 6: lusterng lusterng Groupng data nto (hopefully useful) sets. Thngs on the left Thngs on the rght Applcatons of lusterng Hypothess Generaton lusters mght suggest natural groups. Hypothess
More informationFeature Reduction and Selection
Feature Reducton and Selecton Dr. Shuang LIANG School of Software Engneerng TongJ Unversty Fall, 2012 Today s Topcs Introducton Problems of Dmensonalty Feature Reducton Statstc methods Prncpal Components
More informationParallelism for Nested Loops with Non-uniform and Flow Dependences
Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr
More informationA Background Subtraction for a Vision-based User Interface *
A Background Subtracton for a Vson-based User Interface * Dongpyo Hong and Woontack Woo KJIST U-VR Lab. {dhon wwoo}@kjst.ac.kr Abstract In ths paper, we propose a robust and effcent background subtracton
More informationBAYESIAN MULTI-SOURCE DOMAIN ADAPTATION
BAYESIAN MULTI-SOURCE DOMAIN ADAPTATION SHI-LIANG SUN, HONG-LEI SHI Department of Computer Scence and Technology, East Chna Normal Unversty 500 Dongchuan Road, Shangha 200241, P. R. Chna E-MAIL: slsun@cs.ecnu.edu.cn,
More informationLecture 5: Multilayer Perceptrons
Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented
More informationFast Sparse Gaussian Processes Learning for Man-Made Structure Classification
Fast Sparse Gaussan Processes Learnng for Man-Made Structure Classfcaton Hang Zhou Insttute for Vson Systems Engneerng, Dept Elec. & Comp. Syst. Eng. PO Box 35, Monash Unversty, Clayton, VIC 3800, Australa
More informationSupport Vector Machines
/9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.
More informationAn Evaluation of Divide-and-Combine Strategies for Image Categorization by Multi-Class Support Vector Machines
An Evaluaton of Dvde-and-Combne Strateges for Image Categorzaton by Mult-Class Support Vector Machnes C. Demrkesen¹ and H. Cherf¹, ² 1: Insttue of Scence and Engneerng 2: Faculté des Scences Mrande Galatasaray
More informationSVM-based Learning for Multiple Model Estimation
SVM-based Learnng for Multple Model Estmaton Vladmr Cherkassky and Yunqan Ma Department of Electrcal and Computer Engneerng Unversty of Mnnesota Mnneapols, MN 55455 {cherkass,myq}@ece.umn.edu Abstract:
More informationImage Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline
mage Vsualzaton mage Vsualzaton mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and Analyss outlne mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and
More informationA Fast Visual Tracking Algorithm Based on Circle Pixels Matching
A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng
More informationCollaboratively Regularized Nearest Points for Set Based Recognition
Academc Center for Computng and Meda Studes, Kyoto Unversty Collaboratvely Regularzed Nearest Ponts for Set Based Recognton Yang Wu, Mchhko Mnoh, Masayuk Mukunok Kyoto Unversty 9/1/013 BMVC 013 @ Brstol,
More informationX- Chart Using ANOM Approach
ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are
More informationIEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 25, NO. 4, APRIL
IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 25, NO. 4, APRIL 2016 1713 Weakly Supervsed Fne-Graned Categorzaton Wth Part-Based Image Representaton Yu Zhang, Xu-Shen We, Janxn Wu, Member, IEEE, Janfe Ca,
More informationA Modified Median Filter for the Removal of Impulse Noise Based on the Support Vector Machines
A Modfed Medan Flter for the Removal of Impulse Nose Based on the Support Vector Machnes H. GOMEZ-MORENO, S. MALDONADO-BASCON, F. LOPEZ-FERRERAS, M. UTRILLA- MANSO AND P. GIL-JIMENEZ Departamento de Teoría
More informationGender Classification using Interlaced Derivative Patterns
Gender Classfcaton usng Interlaced Dervatve Patterns Author Shobernejad, Ameneh, Gao, Yongsheng Publshed 2 Conference Ttle Proceedngs of the 2th Internatonal Conference on Pattern Recognton (ICPR 2) DOI
More informationImprovement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration
Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,
More information12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification
Introducton to Artfcal Intellgence V22.0472-001 Fall 2009 Lecture 24: Nearest-Neghbors & Support Vector Machnes Rob Fergus Dept of Computer Scence, Courant Insttute, NYU Sldes from Danel Yeung, John DeNero
More informationDetection of an Object by using Principal Component Analysis
Detecton of an Object by usng Prncpal Component Analyss 1. G. Nagaven, 2. Dr. T. Sreenvasulu Reddy 1. M.Tech, Department of EEE, SVUCE, Trupath, Inda. 2. Assoc. Professor, Department of ECE, SVUCE, Trupath,
More informationFace Recognition Based on SVM and 2DPCA
Vol. 4, o. 3, September, 2011 Face Recognton Based on SVM and 2DPCA Tha Hoang Le, Len Bu Faculty of Informaton Technology, HCMC Unversty of Scence Faculty of Informaton Scences and Engneerng, Unversty
More informationClassification of Face Images Based on Gender using Dimensionality Reduction Techniques and SVM
Classfcaton of Face Images Based on Gender usng Dmensonalty Reducton Technques and SVM Fahm Mannan 260 266 294 School of Computer Scence McGll Unversty Abstract Ths report presents gender classfcaton based
More informationFace Detection with Deep Learning
Face Detecton wth Deep Learnng Yu Shen Yus122@ucsd.edu A13227146 Kuan-We Chen kuc010@ucsd.edu A99045121 Yzhou Hao y3hao@ucsd.edu A98017773 Mn Hsuan Wu mhwu@ucsd.edu A92424998 Abstract The project here
More informationSkew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach
Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research
More informationDiscriminative classifiers for object classification. Last time
Dscrmnatve classfers for object classfcaton Thursday, Nov 12 Krsten Grauman UT Austn Last tme Supervsed classfcaton Loss and rsk, kbayes rule Skn color detecton example Sldng ndo detecton Classfers, boostng
More informationSupport Vector classifiers for Land Cover Classification
Map Inda 2003 Image Processng & Interpretaton Support Vector classfers for Land Cover Classfcaton Mahesh Pal Paul M. Mather Lecturer, department of Cvl engneerng Prof., School of geography Natonal Insttute
More informationSubspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;
Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features
More informationOptimizing Document Scoring for Query Retrieval
Optmzng Document Scorng for Query Retreval Brent Ellwen baellwe@cs.stanford.edu Abstract The goal of ths project was to automate the process of tunng a document query engne. Specfcally, I used machne learnng
More informationWavelets and Support Vector Machines for Texture Classification
Wavelets and Support Vector Machnes for Texture Classfcaton Kashf Mahmood Rapoot Faculty of Computer Scence & Engneerng, Ghulam Ishaq Khan Insttute, Top, PAKISTAN. kmr@gk.edu.pk Nasr Mahmood Rapoot Department
More informationTN348: Openlab Module - Colocalization
TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages
More informationSupport Vector Machine for Remote Sensing image classification
Support Vector Machne for Remote Sensng mage classfcaton Hela Elmanna #*, Mohamed Ans Loghmar #, Mohamed Saber Naceur #3 # Laboratore de Teledetecton et Systeme d nformatons a Reference spatale, Unversty
More informationMULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION
MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION Paulo Quntlano 1 & Antono Santa-Rosa 1 Federal Polce Department, Brasla, Brazl. E-mals: quntlano.pqs@dpf.gov.br and
More informationIncremental Learning with Support Vector Machines and Fuzzy Set Theory
The 25th Workshop on Combnatoral Mathematcs and Computaton Theory Incremental Learnng wth Support Vector Machnes and Fuzzy Set Theory Yu-Mng Chuang 1 and Cha-Hwa Ln 2* 1 Department of Computer Scence and
More informationAn Improved Image Segmentation Algorithm Based on the Otsu Method
3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,
More informationEfficient Segmentation and Classification of Remote Sensing Image Using Local Self Similarity
ISSN(Onlne): 2320-9801 ISSN (Prnt): 2320-9798 Internatonal Journal of Innovatve Research n Computer and Communcaton Engneerng (An ISO 3297: 2007 Certfed Organzaton) Vol.2, Specal Issue 1, March 2014 Proceedngs
More informationNovel Fuzzy logic Based Edge Detection Technique
Novel Fuzzy logc Based Edge Detecton Technque Aborsade, D.O Department of Electroncs Engneerng, adoke Akntola Unversty of Tech., Ogbomoso. Oyo-state. doaborsade@yahoo.com Abstract Ths paper s based on
More informationFace Recognition Method Based on Within-class Clustering SVM
Face Recognton Method Based on Wthn-class Clusterng SVM Yan Wu, Xao Yao and Yng Xa Department of Computer Scence and Engneerng Tong Unversty Shangha, Chna Abstract - A face recognton method based on Wthn-class
More informationDetermining the Optimal Bandwidth Based on Multi-criterion Fusion
Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn
More informationOutline. Type of Machine Learning. Examples of Application. Unsupervised Learning
Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton
More informationHistogram of Template for Pedestrian Detection
PAPER IEICE TRANS. FUNDAMENTALS/COMMUN./ELECTRON./INF. & SYST., VOL. E85-A/B/C/D, No. xx JANUARY 20xx Hstogram of Template for Pedestran Detecton Shaopeng Tang, Non Member, Satosh Goto Fellow Summary In
More informationWishing you all a Total Quality New Year!
Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma
More informationDetection of hand grasping an object from complex background based on machine learning co-occurrence of local image feature
Detecton of hand graspng an object from complex background based on machne learnng co-occurrence of local mage feature Shnya Moroka, Yasuhro Hramoto, Nobutaka Shmada, Tadash Matsuo, Yoshak Shra Rtsumekan
More informationA Statistical Model Selection Strategy Applied to Neural Networks
A Statstcal Model Selecton Strategy Appled to Neural Networks Joaquín Pzarro Elsa Guerrero Pedro L. Galndo joaqun.pzarro@uca.es elsa.guerrero@uca.es pedro.galndo@uca.es Dpto Lenguajes y Sstemas Informátcos
More informationS1 Note. Basis functions.
S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type
More informationLearning Ensemble of Local PDM-based Regressions. Yen Le Computational Biomedicine Lab Advisor: Prof. Ioannis A. Kakadiaris
Learnng Ensemble of Local PDM-based Regressons Yen Le Computatonal Bomedcne Lab Advsor: Prof. Ioanns A. Kakadars 1 Problem statement Fttng a statstcal shape model (PDM) for mage segmentaton Callosum segmentaton
More informationAn Image Fusion Approach Based on Segmentation Region
Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua
More informationBiostatistics 615/815
The E-M Algorthm Bostatstcs 615/815 Lecture 17 Last Lecture: The Smplex Method General method for optmzaton Makes few assumptons about functon Crawls towards mnmum Some recommendatons Multple startng ponts
More informationJournal of Chemical and Pharmaceutical Research, 2014, 6(6): Research Article. A selective ensemble classification method on microarray data
Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(6):2860-2866 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A selectve ensemble classfcaton method on mcroarray
More informationUsing Neural Networks and Support Vector Machines in Data Mining
Usng eural etworks and Support Vector Machnes n Data Mnng RICHARD A. WASIOWSKI Computer Scence Department Calforna State Unversty Domnguez Hlls Carson, CA 90747 USA Abstract: - Multvarate data analyss
More informationFast Feature Value Searching for Face Detection
Vol., No. 2 Computer and Informaton Scence Fast Feature Value Searchng for Face Detecton Yunyang Yan Department of Computer Engneerng Huayn Insttute of Technology Hua an 22300, Chna E-mal: areyyyke@63.com
More informationA Probabilistic Approach to Detect Urban Regions from Remotely Sensed Images Based on Combination of Local Features
A Probablstc Approach to Detect Urban Regons from Remotely Sensed Images Based on Combnaton of Local Features Berl Sırmaçek German Aerospace Center (DLR) Remote Sensng Technology Insttute Weßlng, 82234,
More informationSLAM Summer School 2006 Practical 2: SLAM using Monocular Vision
SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,
More informationEnhanced Face Detection Technique Based on Color Correction Approach and SMQT Features
Journal of Software Engneerng and Applcatons, 2013, 6, 519-525 http://dx.do.org/10.4236/jsea.2013.610062 Publshed Onlne October 2013 (http://www.scrp.org/journal/jsea) 519 Enhanced Face Detecton Technque
More informationCombined Object Detection and Segmentation
Combned Object Detecton and Segmentaton Jarch Vansteenberge, Masayuk Mukunok, and Mchhko Mnoh Abstract We develop a method for combned object detecton and segmentaton n natural scene. In our approach segmentaton
More informationWIRELESS CAPSULE ENDOSCOPY IMAGE CLASSIFICATION BASED ON VECTOR SPARSE CODING.
WIRELESS CAPSULE ENDOSCOPY IMAGE CLASSIFICATION BASED ON VECTOR SPARSE CODING Tao Ma 1, Yuexan Zou 1 *, Zhqang Xang 1, Le L 1 and Y L 1 ADSPLAB/ELIP, School of ECE, Pekng Unversty, Shenzhen 518055, Chna
More informationA Novel Term_Class Relevance Measure for Text Categorization
A Novel Term_Class Relevance Measure for Text Categorzaton D S Guru, Mahamad Suhl Department of Studes n Computer Scence, Unversty of Mysore, Mysore, Inda Abstract: In ths paper, we ntroduce a new measure
More informationUSING LINEAR REGRESSION FOR THE AUTOMATION OF SUPERVISED CLASSIFICATION IN MULTITEMPORAL IMAGES
USING LINEAR REGRESSION FOR THE AUTOMATION OF SUPERVISED CLASSIFICATION IN MULTITEMPORAL IMAGES 1 Fetosa, R.Q., 2 Merelles, M.S.P., 3 Blos, P. A. 1,3 Dept. of Electrcal Engneerng ; Catholc Unversty of
More informationComparison Study of Textural Descriptors for Training Neural Network Classifiers
Comparson Study of Textural Descrptors for Tranng Neural Network Classfers G.D. MAGOULAS (1) S.A. KARKANIS (1) D.A. KARRAS () and M.N. VRAHATIS (3) (1) Department of Informatcs Unversty of Athens GR-157.84
More informationNetwork Intrusion Detection Based on PSO-SVM
TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*
More informationThe Codesign Challenge
ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.
More informationUser Authentication Based On Behavioral Mouse Dynamics Biometrics
User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA
More informationCourse Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms
Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques
More informationVol. 5, No. 3 March 2014 ISSN Journal of Emerging Trends in Computing and Information Sciences CIS Journal. All rights reserved.
Journal of Emergng Trends n Computng and Informaton Scences 009-03 CIS Journal. All rghts reserved. http://www.csjournal.org Unhealthy Detecton n Lvestock Texture Images usng Subsampled Contourlet Transform
More informationCluster Analysis of Electrical Behavior
Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School
More informationCS 534: Computer Vision Model Fitting
CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust
More informationLearning-based License Plate Detection on Edge Features
Learnng-based Lcense Plate Detecton on Edge Features Wng Teng Ho, Woo Hen Yap, Yong Haur Tay Computer Vson and Intellgent Systems (CVIS) Group Unverst Tunku Abdul Rahman, Malaysa wngteng_h@yahoo.com, woohen@yahoo.com,
More informationMachine Learning: Algorithms and Applications
14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of
More informationOnline Detection and Classification of Moving Objects Using Progressively Improving Detectors
Onlne Detecton and Classfcaton of Movng Objects Usng Progressvely Improvng Detectors Omar Javed Saad Al Mubarak Shah Computer Vson Lab School of Computer Scence Unversty of Central Florda Orlando, FL 32816
More informationFace Recognition University at Buffalo CSE666 Lecture Slides Resources:
Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural
More informationImpact of a New Attribute Extraction Algorithm on Web Page Classification
Impact of a New Attrbute Extracton Algorthm on Web Page Classfcaton Gösel Brc, Banu Dr, Yldz Techncal Unversty, Computer Engneerng Department Abstract Ths paper ntroduces a new algorthm for dmensonalty
More informationUnder-Sampling Approaches for Improving Prediction of the Minority Class in an Imbalanced Dataset
Under-Samplng Approaches for Improvng Predcton of the Mnorty Class n an Imbalanced Dataset Show-Jane Yen and Yue-Sh Lee Department of Computer Scence and Informaton Engneerng, Mng Chuan Unversty 5 The-Mng
More informationDeep Classification in Large-scale Text Hierarchies
Deep Classfcaton n Large-scale Text Herarches Gu-Rong Xue Dkan Xng Qang Yang 2 Yong Yu Dept. of Computer Scence and Engneerng Shangha Jao-Tong Unversty {grxue, dkxng, yyu}@apex.sjtu.edu.cn 2 Hong Kong
More informationarxiv: v2 [cs.cv] 9 Apr 2018
Boundary-senstve Network for Portrat Segmentaton Xanzh Du 1, Xaolong Wang 2, Dawe L 2, Jngwen Zhu 2, Serafettn Tasc 2, Cameron Uprght 2, Stephen Walsh 2, Larry Davs 1 1 Computer Vson Lab, UMIACS, Unversty
More informationDiscriminative Dictionary Learning with Pairwise Constraints
Dscrmnatve Dctonary Learnng wth Parwse Constrants Humn Guo Zhuoln Jang LARRY S. DAVIS UNIVERSITY OF MARYLAND Nov. 6 th, Outlne Introducton/motvaton Dctonary Learnng Dscrmnatve Dctonary Learnng wth Parwse
More informationMulticlass Object Recognition based on Texture Linear Genetic Programming
Multclass Object Recognton based on Texture Lnear Genetc Programmng Gustavo Olague 1, Eva Romero 1 Leonardo Trujllo 1, and Br Bhanu 2 1 CICESE, Km. 107 carretera Tjuana-Ensenada, Mexco, olague@ccese.mx,
More informationLearning-Based Top-N Selection Query Evaluation over Relational Databases
Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **
More informationAssociative Based Classification Algorithm For Diabetes Disease Prediction
Internatonal Journal of Engneerng Trends and Technology (IJETT) Volume-41 Number-3 - November 016 Assocatve Based Classfcaton Algorthm For Dabetes Dsease Predcton 1 N. Gnana Deepka, Y.surekha, 3 G.Laltha
More informationA fast algorithm for color image segmentation
Unersty of Wollongong Research Onlne Faculty of Informatcs - Papers (Arche) Faculty of Engneerng and Informaton Scences 006 A fast algorthm for color mage segmentaton L. Dong Unersty of Wollongong, lju@uow.edu.au
More informationA Gradient Difference based Technique for Video Text Detection
A Gradent Dfference based Technque for Vdeo Text Detecton Palaahnakote Shvakumara, Trung Quy Phan and Chew Lm Tan School of Computng, Natonal Unversty of Sngapore {shva, phanquyt, tancl }@comp.nus.edu.sg
More informationA NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION
A NEW FUZZY C-MEANS BASED SEGMENTATION STRATEGY. APPLICATIONS TO LIP REGION IDENTIFICATION Mhaela Gordan *, Constantne Kotropoulos **, Apostolos Georgaks **, Ioanns Ptas ** * Bass of Electroncs Department,
More informationComputer Aided Lytic Bone Metastasis Detection Using Regular CT Images
Computer Aded Lytc Bone Metastass Detecton Usng Regular CT Images Janhua Yao, Stacy D. O Connor, Ronald Summers Dagnostc Radology Department, Clncal Center, Naton Insttutes of Health ABSTRACT Ths paper
More information