A Geometric Approach To Fully Automatic Chromosome Segmentation

Size: px
Start display at page:

Download "A Geometric Approach To Fully Automatic Chromosome Segmentation"

Transcription

1 A Geometrc Approach To Fully Automatc Chromosome Segmentaton Shervn Mnaee ECE Department New York Unversty Brooklyn, New York, USA Mehran Fotouh Computer Engneerng Department Sharf Unversty of Technology Tehran, Iran Babak Hossen Khalaj Electrcal Engneerng Department Sharf Unversty of Technology Tehran, Iran arxv: v5 [cs.cv] 1 Sep 2014 Abstract A fundamental task n human chromosome analyss s chromosome segmentaton. Segmentaton plays an mportant role n chromosome karyotypng. The frst step n segmentaton s to remove ntrusve objects such as stan debrs and other noses. The next step s detecton of touchng and overlappng chromosomes, and the fnal step s separaton of such chromosomes. Common methods for separaton between touchng chromosomes are nteractve and requre human nterventon for correct separaton between touchng and overlappng chromosomes. In ths paper, a geometrc-based method s used for automatc detecton of touchng and overlappng chromosomes and separatng them. The proposed scheme performs segmentaton n two phases. In the frst phase, chromosome clusters are detected usng three geometrc crtera, and n the second phase, chromosome clusters are separated usng a cut-lne. Most of earler methods dd not work properly n case of chromosome clusters that contaned more than two chromosomes. Our method, on the other hand, s qute effcent n separaton of such chromosome clusters. At each step, one separaton wll be performed and ths algorthm s repeated untl all ndvdual chromosomes are separated. Another mportant pont about the proposed method s that t uses the geometrc features of chromosomes whch are ndependent of the type of mages and t can easly be appled to any type of mages such as bnary mages and does not requre multspectral mages as well. We have appled our method to a database contanng 62 touchng and partally overlappng chromosomes and a success rate of 91.9% s acheved. I. INTRODUCTION Chromosome karyotypng s an essental task n cytogenetcs and s usually performed n clncal and cancer cytogenetc labs and can be used n the dagnoss of genetc dsorders. The normal human karyotypes contan 22 pars of autosomal chromosomes and one par of sex chromosomes. Chromosome karyotypng s meant to dentfy and assgn each chromosome n the mage to one of the 24 classes. Chromosome karyotypng has three man steps: pre-processng, segmentaton and classfcaton. Among these steps, chromosome segmentaton s very mportant, snce t affects performance of classfcaton whch s the fnal goal. Chromosome mages may have some defects; they may be bent, they may touch or overlap and ther bands may be spread. In addton, snce touchng and overlappng chromosomes exst n almost every metaphase mage, the soluton of ths problem s vtal. The frst step n analyzng a chromosome mage s segmentaton of chromosomes from the mage background, the man methods used n ths step are based on the evaluaton of a global threshold by means of the Otsu method [1], or on a re-thresholdng scheme [2]. Due to the fact that long chromosomes may touch and overlap, the frst segmentaton step s usually unable to dentfy each chromosome as a sngle object, and presents a number of clusters. So far, attempts have been made to deal wth clusters of touchng (but not overlappng) chromosomes [3], [4], [5], and for clusters of overlappng (but not touchng) chromosomes [6], [7], where both of geometrc and ntensty based features have been used to resolve segmentaton ambgutes. Lerner [8] proposed a method to combne the choce of correct cluster dsentanglement wth the classfcaton stage, resultng n a classfcaton-drven segmentaton. Grsan [9] proposed a smlar method. There are many other methods for separaton between touchng and overlappng objects [10], [11]. Schwartzkopf [12] proposed a method for jont segmentaton and classfcaton that used statstcal method. Snce ths method was appled to multspectral chromosome mages, t does not work for bnary mages. So far, most of chromosome analyss systems have a common fault: ther poor automatc chromosome ncson ablty. Most of current systems for automatc chromosome segmentaton are nteractve and need human nterventon. We have to menton that the orgnal mages are pre-processed and the chromosomes are segmented from the background and the ntrusve objects and noses are removed from the background. Therefore, our man effort s to detect and separate touchng or overlappng chromosomes. It s worth mentonng that there are dfferent approaches for segmentaton and classfcaton of medcal mages. One man approach s to the geometrc characterstcs of the object of nterest, the other one s usng spatal and transform doman features of the mage, etc. The rght set of features depends on the applcaton. For example, n bometrc recognton area, there are a lot of works based on spatal and frequency doman nformaton of mages. One such work s presented n [13], where the author uses the spatal and wavelet doman features of mages to perform palmprnt recognton. However, some of those approaches requres a very large dataset to tran the model properly so they may not be applcable to small datasets, because they could be very prone to over-fttng. A good work for dealng wth small dataset s presented n [14], where the author explans how to jontly maxmze the model

2 accuracy and relablty. In ths paper, a geometrc method for segmentaton of the touchng and partally overlappng chromosomes s presented. Frst, we ntroduce an approach to evaluate whether an object s a sngle chromosome or a chromosome cluster. By chromosome clusters, we mean a group of chromosomes whch overlap and touch each other. Subsequently, for each cluster, we use geometrc features of chromosome boundary whch help separate touchng or partally overlappng chromosomes. Chromosome segmentaton s performed n two phases. In the frst phase, touchng or overlappng chromosomes are detected usng the approach whch s ntroduced n Secton II where we deal wth the chromosomes shape and ther geometrc features. If two or more chromosomes overlap, the resultng cluster would not have the usual long and thn shape and we can use such dfference to detect chromosome clusters. In the second phase, we use other geometrc features to separate touchng or partally overlappng chromosomes. We wll dscuss about ths step n Secton III whch deals wth boundary pattern of chromosomes. Our method has three advantages over earler schemes: 1) Frst, t can be appled to any type of mages, even bnary mage, and t does not need multspectral or grayscale mages. Therefore t can reduce the cost of photography and the amount of computaton. 2) Second, t can easly separate chromosome clusters that contan more than two chromosomes where most earler schemes fal. 3) Thrd, our method s fully automatc and does not need any human nterventon. II. DETECTION OF TOUCHING OR OVERLAPPING CHROMOSOMES In order to detect chromosome clusters, we use three crtera whch deal wth the geometry of chromosomes. The frst method s surroundng ellpse method (Secton II.A), whch s based on the rato of the length of mnor axs of surroundng ellpse to the length of ts major axs. The second method s convex hull method (Secton II.B) whch s based on the number of pxels n the orgnal chromosome to the number of pxels n ts convex hull rato. An mportant pont about ths method s ts robustness n detectng small chromosomes that may produce error n the frst method. The thrd method s skeleton and end ponts (Secton II.C), whch uses the skeleton of each chromosome (ether sngle chromosome or a chromosome cluster) to fnd the end ponts of skeleton and decdes based on the number of end ponts. All of these methods have some lmtatons, but through proper ntegraton, we can detect all chromosome clusters (ether touchng or overlappng chromosomes) as shown through our smulaton results. Each chromosome passes through these three methods and n case t satsfes the crtera of all three methods, t wll be detected as a chromosome cluster. We wll dscuss the detals of each method n the followng parts. A. Surroundng ellpse method Surroundng ellpse of a shape s an ellpse whch surrounds that. Sngle chromosome s usually long and thn (unless those chromosomes whch belong to 20th, 21st or 22nd group) so ts surroundng ellpse wll be long, but the overlappng chromosomes have a surroundng ellpse close to a crcle. We can use ths dfference for detecton of overlappng chromosomes. In order to take advantage of such dfference, the rato of the length of mnor to major axs of the surroundng ellpse has to be found. If the label s overlappng, we expect ths rato to be close to 1, because the surroundng ellpse would be close to a crcle, but f the chromosome s sngle t wll have a smaller rato. Therefore, a threshold can be determned to dstngush between chromosome clusters and sngle chromosomes. We propose the below algorthm for ths step: 1) Fnd the surroundng ellpse of each chromosome label 2) Fnd the rato of mnor axs length to major axs length of surroundng ellpse for each chromosome (ether a sngle chromosome or chromosome cluster) 3) Determne a threshold (we smply set ths threshold to the average of all ratos, but f we have a large dataset, we can use a tranng set to determne ths threshold) 4) Compare rato of each label wth ths threshold. For each label, f the rato s less than the threshold, remove t, but f the rato s more than the threshold, keep ths chromosome Ths method s very fast, but t has problems wth two types of chromosomes: 1) Small chromosomes 2) Bent chromosomes The shape of small chromosomes s dfferent from usual chromosomes as they have a round shape where the rato of mnor axs length to major axs length of ther surroundng ellpse wll be smlar to overlappng chromosomes. Bent chromosomes also have ratos smlar to overlappng chromosomes so they may wrongly be detected as overlappng chromosomes, an ssue that needs to be addressed properly. As the proposed algorthms are appled to each chromosome n a cascade fashon, each step has to remove those sngle chromosomes whch are not removed n the prevous steps. B. Convex hull method In Eucldean space, an object s convex f for every par of ponts wthn the object, every pont on the straght lne segment that jons them s also wthn the object. The convex hull of a set C s the smallest convex set that contans C. Convex hull have been used n several applcatons n computer vson, mage analyss, and dgtal mage processng, ncludng object recognton, mage and vdeo codng. As a normal chromosome has a relatvely convex shape, ts convex hull would approxmately have the same number of pxels as the orgnal chromosome. If we fnd the convex hull of the chromosomes, we wll notce that the convex hull of chromosome clusters have much more pxels than chromosome clusters themselves, whereas the sngle chromosomes have almost the

3 same number of pxels as ther convex hulls. Consequently, we can detect chromosome clusters usng such dfference. In order to acheve ths goal, we should fnd the rato of the number of pxels n each chromosome to number of pxel n ts convex hull for all chromosomes and then compare these ratos wth a threshold. For each chromosome, f the rato s less than a gven threshold, we expect that ths label would be an overlappng chromosomes and vce versa. The proposed algorthm for ths method s gven below: 1) Fnd convex hull of each chromosome label. 2) Calculate the rato of the number of pxels n the orgnal chromosome to the number of pxel n ts convex hull for each chromosome. 3) Determne a threshold (ths threshold can be determned usng tranng set, or t can smply set to the average of these ratos for all chromosomes) 4) Compare ths rato for each chromosome wth ths threshold, for each label f the rato was more than the threshold elmnate ths chromosomes. 5) The remanng chromosomes wll be sent to the next step. One advantage of ths method s that we can elmnate small sngle chromosomes remanng from the prevous step. Snce for these chromosomes the convex hull s almost concdent wth orgnal chromosome, the rato wll be more than the gven threshold. However, as we stll have problem wth bent chromosomes, we should elmnate them n the next step. Fg.1 represents an mage of chromosomes wth convex hulls of two chromosomes. Fg. 1. A chromosome mage wth convex hull of two chromosomes C. Skeleton and end ponts Skeletonzaton s the transformaton of a component n a dgtal mage nto a subset of the orgnal component. Skeleton has been used n several applcatons n computer vson, mage analyss, and dgtal mage processng. We used ths method as one step of the chromosome clusters detecton algorthm. If we fnd the skeleton of each chromosome and then fnd the end ponts of ths skeleton (end ponts are those pont whch are the last pont n any sde of a lne) we wll notce that the overlappng chromosomes usually have more than 2 end ponts. Therefore, we can use ths dea for detecton of overlappng chromosomes. Skeletons and end ponts of a set of chromosomes are represented n Fg.2. End ponts of chromosomes are shown wth red ponts. We observe that all chromosomes clusters n ths pcture have more than two end ponts. Fg. 2. An example of skeletons and end ponts of a chromosome mage The proposed algorthm for ths method s: 1) Fnd the skeleton of each chromosomes. 2) Fnd the end ponts of each skeleton. 3) In the case more than two ponts are found, classfy them as a chromosome clusters. Ths method s robust for fndng overlappng chromosomes. However, because of the teratve structure of the skeleton algorthm, t s tme-consumng and we should mprove ts computatonal complexty. In the next part, we combne these three methods n a proper way. D. Integraton Step In order to solve the problem of the skeleton method, we decded to apply ths method to a fewer number of chromosomes, ntally 40 to 46 chromosomes, some of whch are overlappng. Frst, we apply convex hull and surroundng ellpse methods and elmnate a large number of sngle chromosomes. After these two steps we usually have about 8 to 14 chromosomes. Subsequently, the skeleton method can be appled to detect chromosome clusters from the remanng ones and because the number of chromosomes has been reduced from 46 to between 8 and 14, we wll mprove the tme effcency of the algorthm by a factor of 4. On the other hand, the surroundng ellpse method can not elmnate small sngle chromosomes and t s better to apply surroundng ellpse method after convex hull method. The block dagram for the drecton of overlappng chromosomes shown n Fg.3 : Each chromosome passes through these three methods and f t satsfes all three crtera, t wll be consdered as a chromosome cluster. After detecton of all chromosome clusters, they wll be used as the nput of the second phase, whch s separaton of chromosome clusters.

4 Fg. 3. Block dagram of chromosome clusters detecton algorthm III. SEPARATION OF TOUCHING OR PARTIALLY OVERLAPPING CHROMOSOMES USING CUT-LINE METHOD As dscussed prevously, after detecton of overlappng chromosomes we need to separate them. We ntroduce another geometrc-based method for separaton of touchng or partally overlappng chromosomes. Frst, we fnd the cross-ponts of overlappng chromosomes. Crossponts are those ponts on the boundary of a chromosome cluster where two chromosomes touch or overlap. We wll dscuss the methods whch can be used to fnd these crossponts n Sectons III.A and III.B. Through the applcaton of the proposed method, varous touchng or partally overlappng chromosomes can be handled n the same way. Two chromosome clusters and ther cross-ponts are shown n Fg.4: Fg. 4. Cross-ponts of chromosome clusters Once the cross-ponts are found, we should separate chromosome clusters usng these cross-ponts. If the chromosome cluster conssts of two chromosomes, t can then be cut from the lne between cross-ponts resultng n two sngle chromosomes. However, f t conssts of more than two chromosomes, we should repeat the whole algorthm multple tmes. In the followng sectons, we wll ntroduce some approaches to fnd such cross-ponts. In order to fnd these cross-ponts, we only need to search on the boundary of the chromosomes. Therefore, n order to reduce the amount of computatons, we can extract the boundary of chromosomes and search for the cross-ponts only n the boundary locatons. Once the boundary extracton s done, we can sort the pxels on the boundary n a clockwse fashon. Suppose that the sorted boundary pxels are located n an N 2 matrx n whch each row contans the coordnates of the -th pxel on the boundary and N s the total number of pxels on the boundary. Ths boundary matrx wll be denoted by B. Fg.5 llustrates the result of boundary extracton n a chromosome cluster. In order to fnd the cross-ponts, we use two crtera based on the geometry of the boundary. The crtera are: 1) Varatons n the Angle of Moton Drecton (VAMD) 2) Sum of Dstances among Total Ponts (SDTP) Fg. 5. A chromosome cluster and ts boundary The frst crteron, VAMD, s explaned n Secton III.A. It tres to fnd the cross-ponts based on the varaton n the angle of moton. The second crteron s SDTP as explaned n Secton III.B. Ths crteron uses the fact that cross-ponts are usually located n the mddle of a chromosome cluster. All pxels of the boundary pass through these two crtera, and at each step, some of boundary ponts wll be elmnated and the total crossponts wll be selected wth a cost functon whch takes nto account both these crtera. A. Varatons n the Angle of Moton Drecton (VAMD) In order to understand the meanng of VAMD, suppose that an object s movng on the boundary of a chromosome. At each pxel, t has to move n a drecton called moton drecton whch leads t to the next pxel. The angle between ths drecton and the horzontal axs s called the angle of moton drecton. Ths angle can be calculated as the angle of connectng lne between -th and (+1)-th pxels. We denote ths angle wth θ. ( ) y(+1) y() θ = tan 1 x(+1) x() It should be noted that due to nose on the boundary, t s better to use more pxels to fnd a better estmaton of the angle of moton drecton. One can use angle of the connectng lne between -th pxel and j-th pxel as: θ (j) ( ) y(j) y() = tan 1 x(j) x() Subsequently, we can use a weghted average of θ (j) dfferent j s to fnd a robust estmaton of θ. θ = +N1 j=+1 β(j) θ (j) j= N 2 α (j) θ (j) + 1 +N1 j=+1 β(j) + 1 j= N 2 α (j) usng the forthcomng pxels and the second summaton s the estmated θ (j) and α (j) can be set to The frst summaton s the estmated θ (j) usng prevous pxels. The weghts β (j) a fxed value or can be adaptve. The adaptve choce usually works better and t has to be a functon of the Eucldean dstance between the -th and j-th pxels. For example, one possble choce of β (j) and α (j) could be e (d(b,bj))2, where d(b,b j ) denotes the Eucldean dstance between -th and j-th (1) (2) for (3)

5 pxels. As can be verfed from the above formula, for pxels wth long dstance from the current pxel, d(b,b j ) would be large, therefore the correspondng weghts would be very small, whch s reasonable. Based on our smulaton we deduced that f j=+4, +5 are used, the result wll be the most satsfactory. Therefore, we used the followng formula to fnd θ : θ = (1/2) (θ (+4) Fg.6 depcts ths method on a curve. Fg. 6. +θ (+5) ) (4) Representaton of boundary pxel s angle After fndng θ for all pxels on the boundary, we have to calculate the varaton of angle n the -th pxel as the dfference of the (+1)-th pxel angle and the -th pxel. θ = θ +1 θ (5) We expect to have a larger θ n cross-ponts compared to the other ponts of the boundary. We can use the followng algorthm to remove superfluous ponts on the boundary: 1) Fnd the varaton of angle n each pxel on the boundary. N =1 2) Calculate the average of θ : θ avg = θ N. 3) For each pxel f θ < λ 1 θ avg then remove ths pxel from canddate pxels for cross-ponts (we used λ 1 =1 whch s found by tral and error). The result of ths step s shown n Fg.7. n the cross-ponts to be less than the other ponts. We can fnd the sum of dstances from each pxel to other pxels as: Ds() = M d(b,b j ) for = 1 : M (6) j=1 j where d(b,b j ) s the Eucldean dstance between -th and j-th pxels on the boundary. We can select two pxels wth the mnmum amount of Ds() as cross-ponts. However n some cases, ths method can select the wrong ponts. For example, f one small chromosome touches a large chromosome by ts end, ths selecton method wll not work properly. Therefore, we have to use another crteron alongsde ths crteron n order to avod such errors. In order to avod ms-selecton, we can use both VAMD and SDTP n our fnal decson crteron. Therefore, n cases that SDPT cannot select the rght ponts, VAMD can help the algorthm avod ms-selecton. We defned a cost functon whch takes nto account both VAMD and SDTP and selects two ponts wth mnmum amount of cost functon as the crossponts. Cost() = Ds() λ θ (7) The parameter λ should be a postve number that can control the effect of θ n the cost functon. In order to mnmze ths functon, one needs to mnmze Ds() and maxmze θ. The value of λ can be determned by tral and error on a tranng set. Based on our smulaton λ=1000 produces satsfactory results. After applyng ths algorthm, we wll choose two ponts wth the two least values of Cost() as the cross-ponts and overlappng chromosome can be separated usng the lne between these two ponts. The result of ths step s shown n Fg.8. Fg. 7. Remanng pxels after applyng VAMD crteron Fg. 8. Separaton of chromosome clusters wth proposed method B. Sum of Dstances among Total Ponts (SDTP) Because of nose n chromosomes mages, after the aforementoned step, there may reman more than two ponts, so we should use another crteron to fnd the cross-ponts. Let us assume after the prevous step, M ponts have remaned. We denote these ponts wth B 1 to B M where M s the number of remanng ponts. The cross-ponts are usually located n the mddle of overlappng chromosomes (because overlappng chromosomes are formed by two or more chromosomes). For each remanng pxel, we fnd the sum of dstances between ths pxel and other remanng pxels and we expect ths sum IV. SEPARATION OF CHROMOSOME CLUSTERS WITH MORE THAN TWO CHROMOSOMES In the prevous sectons, we concentrated on chromosome clusters consstng of two chromosomes. In some cases, chromosome clusters may have more than two chromosomes. For bnary mages, t s dffcult to separate a chromosome cluster wth more than two sngle chromosomes n one step. However, we can easly separate ths type of chromosome clusters by a mult-step algorthm. In fact, f a chromosome cluster conssts of N sngle chromosomes, we can separate all sngle chromosomes n N-1 steps. We propose the followng algorthm for separaton of clusters wth more than two chromosomes:

6 TABLE I PROPOSED METHOD ACCURACY, COMMPARED TO METHODS IN THE LITERATURE Method Number of touchng or partally overlappng chromosoms Accuracy J (1989) [1] set % J (1989) [1] set % Lerner (1998) [4] 46 82% Grsan (2009) [9] % Proposed method % 1) Separate each chromosome cluster wth prevous methods. After separaton we wll have two new chromosomes. 2) For each new chromosome check whether t s a sngle chromosome or a chromosome cluster. 3) If both of the new chromosomes are sngle, the algorthm s fnshed. 4) If at least one of the new chromosome s a chromosome cluster, separate t usng the procedure n the second phase. 5) Contnue ths algorthm untl all new chromosomes are sngle. Fg. 9. method Examples of separaton of chromosome clusters usng proposed V. RESULTS By usng the proposed algorthm, we analyzed 25 mages contanng a total number of 1150 chromosomes. There are about 62 touchng or partally overlappng chromosomes n ths data set. We are nterested n assessment of the ablty of the proposed algorthm to successfully separate clusters nto ther composng chromosomes. We tested our algorthm on these 62 chromosome clusters and observed that ths algorthm separates 57 chromosome clusters correctly. So an accuracy rate about 91.9% s attaned. In Table I, we have reported the fracton of correct separatons of touchng and partally overlappng wth respect to ther total number. In ths table, we also have reported a comparson wth all smlar results reported n the lterature. Fg.9 presents results of separaton between sx touchng and partally overlappng chromosomes. As we can see, ths method provdes very effcent separaton of touchng and partally overlappng chromosomes. VI. CONCLUSION In ths paper, a geometrc-based approach s proposed for chromosome segmentaton. It uses three crtera for detecton of chromosome clusters. After that, t uses a novel geometrc method to fnd two cross-ponts on the boundary of clusters whch can be used for extracton of cut-lne. After the cutlne s found, we can decompose groups of chromosomes whch touch and overlap each other. Ths algorthm s able to decompose clusters of touchng or partally overlappng chromosomes that consst of more than two chromosomes. Another advantage of ths method s that t can easly apply to any type of mages, even bnary chromosome mages. In addton, due to use of geometrc features of chromosomes whch are ndependent of mage type, the proposed scheme does not need multspectral mages. In future, we wll focus on separaton of completely overlappng chromosomes. For ths purpose, frst we should dstngush between touchng chromosomes and overlappng chromosomes and apply the related algorthm to each class (separaton algorthm for touchng or partally overlappng chromosomes s dfferent from separaton algorthm for completely overlappng chromosomes). Separaton algorthm for completely overlappng chromosomes s based on fndng the cross secton between two overlappng chromosomes and usng t for separaton of chromosome clusters. ACKNOWLEDGMENTS The authors would lke to thank Prof. Hamd Aghajan for hs nvaluable help durng ths project. We would also thank Mr. Payam Delgosha, Mr. Amral Abdolrashd and Mr. Al Hashem for ther useful comments on our project. REFERENCES [1] L. J, "Intellgent splttng n the chromosome doman," Pattern Recognton, vol. 22, no. 5, pp , [2] L. J, "Fully automatc chromosome segmentaton," Cytometry, vol. 17, pp , [3] B. Lerner, "Toward a completely automatc neural-network-based human chromosome analyss," IEEE Trans. Syst., Man, Cybern. B, Cybern., vol. 28, no. 4, pp , Aug [4] B. Lerner, H. Guterman, and I. Dnsten, "A classfcaton-drven partally occluded object segmentaton (CPOOS) method wth applcaton to chromosome analyss," IEEE Trans. Sgnal Process., vol. 46, no. 10, pp , Oct [5] X. Shunren, X. Wedong, and S. Yutang, "Two ntellgent algorthms appled to automatc chromosome ncson," n Proc. IEEE Int. Conf. Acoust., Speech, Sgnal Process., 03. (ICASSP 03)., Apr., pp

7 [6] G. Agam and I. Dnsten, "Geometrc separaton of partally overlappng nonrgd objects appled to automatc chromosome segmentaton," IEEE Trans. Pattern Anal. Mach. Intell., vol. 19, no. 11, pp , Nov [7] C. Urdales Garca, A. Bandera Rubo, F. Arrebola PÂt erez, and F. Sandoval HernÂt andez, "A curvature-based multresoluton automatc karyotypng system," Mach. Vs. Appl., vol. 14, pp , [8] B. Lerner, H. Guterman, and I. Dnsten, "A classfcaton-drven partally occluded object segmentaton (CPOOS) method wth applcaton to chromosome analyss," IEEE Trans. Sgnal Process., vol. 46, no. 10, pp , Oct [9] Enrco Grsan, Enea Polett, and Alfredo Rugger," Automatc Segmentaton and Dsentanglng of Chromosomes n Q-Band Prometaphase Images" IEEE Trans. Informaton Technology n Bomedcne, vol. 13, no. 4, July [10] M.W.Koch and R.L. Kashyap, "Usng Polygons to Recognze and Locate Partally Occluded Objects," IEEE Trans. Pattern Analyss and Machne Intellgence, vol. 9, no. 4, pp , Apr [11] H.J. Wolfson and Y. Lamdan, "Transformaton Invarant Indexng," Geometrc Invarance n Computer Vson, J.L. Mundy and A. Zsserman, eds., pp MIT Press, [12] Wade C. Schwartzkopf, Alan C. Bovk, and Bran L. Evans," Maxmum- Lkelhood Technques for Jont Segmentaton-Classfcaton of Multspectral Chromosome Images." IEEE Trans on medcal magng, vol. 24, no. 12, Dec 2005 [13] S. Mnaee and A. Abdolrashd. "Hghly Accurate Multspectral Palmprnt Recognton Usng Statstcal and Wavelet Features." arxv preprnt arxv: (2014). [14] S. Mnaee, Y. Wang and Y. W. Lu, "Predcton of Longterm Outcome of Neuropsychologcal Tests of MTBI Patents Usng Imagng Features," Sgnal Processng n Medcne and Bology Symposum (SPMB), IEEE, 2013.

Study and Comparison of Various Techniques of Image Edge Detection

Study and Comparison of Various Techniques of Image Edge Detection Gureet Sngh et al Int. Journal of Engneerng Research Applcatons RESEARCH ARTICLE OPEN ACCESS Study Comparson of Varous Technques of Image Edge Detecton Gureet Sngh*, Er. Harnder sngh** *(Department of

More information

Copy Number Variation Methods and Data

Copy Number Variation Methods and Data Copy Number Varaton Methods and Data Copy number varaton (CNV) Reference Sequence ACCTGCAATGAT TAAGCCCGGG TTGCAACGTTAGGCA Populaton ACCTGCAATGAT TAAGCCCGGG TTGCAACGTTAGGCA ACCTGCAATGAT TTGCAACGTTAGGCA

More information

Proceedings of the 6th WSEAS Int. Conf. on EVOLUTIONARY COMPUTING, Lisbon, Portugal, June 16-18, 2005 (pp )

Proceedings of the 6th WSEAS Int. Conf. on EVOLUTIONARY COMPUTING, Lisbon, Portugal, June 16-18, 2005 (pp ) Proceedngs of the 6th WSEAS Int. Conf. on EVOLUTIONARY COMPUTING, Lsbon, Portugal, June 6-8, 2005 (pp285-20) Novel Intellgent Edge Detector for Sonographcal Images Al Rafee *, Mohammad Hasan Morad **,

More information

A New Machine Learning Algorithm for Breast and Pectoral Muscle Segmentation

A New Machine Learning Algorithm for Breast and Pectoral Muscle Segmentation Avalable onlne www.ejaet.com European Journal of Advances n Engneerng and Technology, 2015, 2(1): 21-29 Research Artcle ISSN: 2394-658X A New Machne Learnng Algorthm for Breast and Pectoral Muscle Segmentaton

More information

*VALLIAPPAN Raman 1, PUTRA Sumari 2 and MANDAVA Rajeswari 3. George town, Penang 11800, Malaysia. George town, Penang 11800, Malaysia

*VALLIAPPAN Raman 1, PUTRA Sumari 2 and MANDAVA Rajeswari 3. George town, Penang 11800, Malaysia. George town, Penang 11800, Malaysia 38 A Theoretcal Methodology and Prototype Implementaton for Detecton Segmentaton Classfcaton of Dgtal Mammogram Tumor by Machne Learnng and Problem Solvng *VALLIAPPA Raman, PUTRA Sumar 2 and MADAVA Rajeswar

More information

Parameter Estimates of a Random Regression Test Day Model for First Three Lactation Somatic Cell Scores

Parameter Estimates of a Random Regression Test Day Model for First Three Lactation Somatic Cell Scores Parameter Estmates of a Random Regresson Test Day Model for Frst Three actaton Somatc Cell Scores Z. u, F. Renhardt and R. Reents Unted Datasystems for Anmal Producton (VIT), Hedeweg 1, D-27280 Verden,

More information

FAST DETECTION OF MASSES IN MAMMOGRAMS WITH DIFFICULT CASE EXCLUSION

FAST DETECTION OF MASSES IN MAMMOGRAMS WITH DIFFICULT CASE EXCLUSION computng@tanet.edu.te.ua www.tanet.edu.te.ua/computng ISSN 727-6209 Internatonal Scentfc Journal of Computng FAST DETECTION OF MASSES IN MAMMOGRAMS WITH DIFFICULT CASE EXCLUSION Gábor Takács ), Béla Patak

More information

A New Diagnosis Loseless Compression Method for Digital Mammography Based on Multiple Arbitrary Shape ROIs Coding Framework

A New Diagnosis Loseless Compression Method for Digital Mammography Based on Multiple Arbitrary Shape ROIs Coding Framework I.J.Modern Educaton and Computer Scence, 2011, 5, 33-39 Publshed Onlne August 2011 n MECS (http://www.mecs-press.org/) A New Dagnoss Loseless Compresson Method for Dgtal Mammography Based on Multple Arbtrary

More information

AUTOMATED DETECTION OF HARD EXUDATES IN FUNDUS IMAGES USING IMPROVED OTSU THRESHOLDING AND SVM

AUTOMATED DETECTION OF HARD EXUDATES IN FUNDUS IMAGES USING IMPROVED OTSU THRESHOLDING AND SVM AUTOMATED DETECTION OF HARD EXUDATES IN FUNDUS IMAGES USING IMPROVED OTSU THRESHOLDING AND SVM Wewe Gao 1 and Jng Zuo 2 1 College of Mechancal Engneerng, Shangha Unversty of Engneerng Scence, Shangha,

More information

Using the Perpendicular Distance to the Nearest Fracture as a Proxy for Conventional Fracture Spacing Measures

Using the Perpendicular Distance to the Nearest Fracture as a Proxy for Conventional Fracture Spacing Measures Usng the Perpendcular Dstance to the Nearest Fracture as a Proxy for Conventonal Fracture Spacng Measures Erc B. Nven and Clayton V. Deutsch Dscrete fracture network smulaton ams to reproduce dstrbutons

More information

Research Article Statistical Analysis of Haralick Texture Features to Discriminate Lung Abnormalities

Research Article Statistical Analysis of Haralick Texture Features to Discriminate Lung Abnormalities Hndaw Publshng Corporaton Internatonal Journal of Bomedcal Imagng Volume 2015, Artcle ID 267807, 7 pages http://dx.do.org/10.1155/2015/267807 Research Artcle Statstcal Analyss of Haralck Texture Features

More information

A Computer-aided System for Discriminating Normal from Cancerous Regions in IHC Liver Cancer Tissue Images Using K-means Clustering*

A Computer-aided System for Discriminating Normal from Cancerous Regions in IHC Liver Cancer Tissue Images Using K-means Clustering* A Computer-aded System for Dscrmnatng Normal from Cancerous Regons n IHC Lver Cancer Tssue Images Usng K-means Clusterng* R. M. CHEN 1, Y. J. WU, S. R. JHUANG, M. H. HSIEH, C. L. KUO, Y. L. MA Department

More information

Physical Model for the Evolution of the Genetic Code

Physical Model for the Evolution of the Genetic Code Physcal Model for the Evoluton of the Genetc Code Tatsuro Yamashta Osamu Narkyo Department of Physcs, Kyushu Unversty, Fukuoka 8-856, Japan Abstract We propose a physcal model to descrbe the mechansms

More information

Shape-based Retrieval of Heart Sounds for Disease Similarity Detection Tanveer Syeda-Mahmood, Fei Wang

Shape-based Retrieval of Heart Sounds for Disease Similarity Detection Tanveer Syeda-Mahmood, Fei Wang Shape-based Retreval of Heart Sounds for Dsease Smlarty Detecton Tanveer Syeda-Mahmood, Fe Wang 1 IBM Almaden Research Center, 650 Harry Road, San Jose, CA 95120. {stf,wangfe}@almaden.bm.com Abstract.

More information

Using a Wavelet Representation for Classification of Movement in Bed

Using a Wavelet Representation for Classification of Movement in Bed Usng a Wavelet Representaton for Classfcaton of Movement n Bed Adrana Morell Adam Depto. de Matemátca e Estatístca Unversdade de Caxas do Sul Caxas do Sul RS E-mal: amorell@ucs.br André Gustavo Adam Depto.

More information

Towards Automated Pose Invariant 3D Dental Biometrics

Towards Automated Pose Invariant 3D Dental Biometrics Towards Automated Pose Invarant 3D Dental Bometrcs Xn ZHONG 1, Depng YU 1, Kelvn W C FOONG, Terence SIM 3, Yoke San WONG 1 and Ho-lun CHENG 3 1. Mechancal Engneerng, Natonal Unversty of Sngapore, 117576,

More information

Journal of Engineering Science and Technology Review 11 (2) (2018) Research Article

Journal of Engineering Science and Technology Review 11 (2) (2018) Research Article Jestr Journal of Engneerng Scence and Technology Revew 11 (2) (2018) 8-12 Research Artcle Detecton Lung Cancer Usng Gray Level Co-Occurrence Matrx (GLCM) and Back Propagaton Neural Network Classfcaton

More information

Biomarker Selection from Gene Expression Data for Tumour Categorization Using Bat Algorithm

Biomarker Selection from Gene Expression Data for Tumour Categorization Using Bat Algorithm Receved: March 20, 2017 401 Bomarker Selecton from Gene Expresson Data for Tumour Categorzaton Usng Bat Algorthm Gunavath Chellamuthu 1 *, Premalatha Kandasamy 2, Svasubramanan Kanagaraj 3 1 School of

More information

Available online at ScienceDirect. Procedia Computer Science 46 (2015 )

Available online at   ScienceDirect. Procedia Computer Science 46 (2015 ) Avalable onlne at www.scencedrect.com ScenceDrect Proceda Computer Scence 46 (215 ) 1762 1769 Internatonal Conference on Informaton and Communcaton Technologes (ICICT 214) Automatc Characterzaton of Bengn

More information

Fast Algorithm for Vectorcardiogram and Interbeat Intervals Analysis: Application for Premature Ventricular Contractions Classification

Fast Algorithm for Vectorcardiogram and Interbeat Intervals Analysis: Application for Premature Ventricular Contractions Classification Fast Algorthm for Vectorcardogram and Interbeat Intervals Analyss: Applcaton for Premature Ventrcular Contractons Classfcaton Irena Jekova, Vessela Krasteva Centre of Bomedcal Engneerng Prof. Ivan Daskalov

More information

IDENTIFICATION AND DELINEATION OF QRS COMPLEXES IN ELECTROCARDIOGRAM USING FUZZY C-MEANS ALGORITHM

IDENTIFICATION AND DELINEATION OF QRS COMPLEXES IN ELECTROCARDIOGRAM USING FUZZY C-MEANS ALGORITHM IDENTIFICATION AND DELINEATION OF QRS COMPLEXES IN ELECTROCARDIOGRAM USING FUZZY C-MEANS ALGORITHM S.S. MEHTA 1, C.R.TRIVEDI 2, N.S. LINGAYAT 3 1 Electrcal Engneerng Department, J.N.V, Unversty, Jodhpur.

More information

Using Past Queries for Resource Selection in Distributed Information Retrieval

Using Past Queries for Resource Selection in Distributed Information Retrieval Purdue Unversty Purdue e-pubs Department of Computer Scence Techncal Reports Department of Computer Scence 2011 Usng Past Queres for Resource Selecton n Dstrbuted Informaton Retreval Sulleyman Cetntas

More information

CLUSTERING is always popular in modern technology

CLUSTERING is always popular in modern technology Max-Entropy Feed-Forward Clusterng Neural Network Han Xao, Xaoyan Zhu arxv:1506.03623v1 [cs.lg] 11 Jun 2015 Abstract The outputs of non-lnear feed-forward neural network are postve, whch could be treated

More information

310 Int'l Conf. Par. and Dist. Proc. Tech. and Appl. PDPTA'16

310 Int'l Conf. Par. and Dist. Proc. Tech. and Appl. PDPTA'16 310 Int'l Conf. Par. and Dst. Proc. Tech. and Appl. PDPTA'16 Akra Sasatan and Hrosh Ish Graduate School of Informaton and Telecommuncaton Engneerng, Toka Unversty, Mnato, Tokyo, Japan Abstract The end-to-end

More information

Improvement of Automatic Hemorrhages Detection Methods using Brightness Correction on Fundus Images

Improvement of Automatic Hemorrhages Detection Methods using Brightness Correction on Fundus Images Improvement of Automatc Hemorrhages Detecton Methods usng Brghtness Correcton on Fundus Images Yuj Hatanaka *a, Toshak Nakagawa *b, Yoshnor Hayash *c, Masakatsu Kakogawa *c, Akra Sawada *d, Kazuhde Kawase

More information

EXAMINATION OF THE DENSITY OF SEMEN AND ANALYSIS OF SPERM CELL MOVEMENT. 1. INTRODUCTION

EXAMINATION OF THE DENSITY OF SEMEN AND ANALYSIS OF SPERM CELL MOVEMENT. 1. INTRODUCTION JOURNAL OF MEDICAL INFORMATICS & TECHNOLOGIES Vol.3/00, ISSN 64-6037 Łukasz WITKOWSKI * mage enhancement, mage analyss, semen, sperm cell, cell moblty EXAMINATION OF THE DENSITY OF SEMEN AND ANALYSIS OF

More information

AUTOMATED CHARACTERIZATION OF ESOPHAGEAL AND SEVERELY INJURED VOICES BY MEANS OF ACOUSTIC PARAMETERS

AUTOMATED CHARACTERIZATION OF ESOPHAGEAL AND SEVERELY INJURED VOICES BY MEANS OF ACOUSTIC PARAMETERS AUTOMATED CHARACTERIZATIO OF ESOPHAGEAL AD SEVERELY IJURED VOICES BY MEAS OF ACOUSTIC PARAMETERS B. García, I. Ruz, A. Méndez, J. Vcente, and M. Mendezona Department of Telecommuncaton, Unversty of Deusto

More information

Dr.S.Sumathi 1, Mrs.V.Agalya 2 Mahendra Engineering College, Mahendhirapuri, Mallasamudram

Dr.S.Sumathi 1, Mrs.V.Agalya 2 Mahendra Engineering College, Mahendhirapuri, Mallasamudram Detecton Of Myocardal Ischema In ECG Sgnals Usng Support Vector Machne Dr.S.Sumath 1, Mrs.V.Agalya Mahendra Engneerng College, Mahendhrapur, Mallasamudram Abstract--Ths paper presents an ntellectual dagnoss

More information

Prediction of Total Pressure Drop in Stenotic Coronary Arteries with Their Geometric Parameters

Prediction of Total Pressure Drop in Stenotic Coronary Arteries with Their Geometric Parameters Tenth Internatonal Conference on Computatonal Flud Dynamcs (ICCFD10), Barcelona, Span, July 9-13, 2018 ICCFD10-227 Predcton of Total Pressure Drop n Stenotc Coronary Arteres wth Ther Geometrc Parameters

More information

Optimal Planning of Charging Station for Phased Electric Vehicle *

Optimal Planning of Charging Station for Phased Electric Vehicle * Energy and Power Engneerng, 2013, 5, 1393-1397 do:10.4236/epe.2013.54b264 Publshed Onlne July 2013 (http://www.scrp.org/ournal/epe) Optmal Plannng of Chargng Staton for Phased Electrc Vehcle * Yang Gao,

More information

Appendix for. Institutions and Behavior: Experimental Evidence on the Effects of Democracy

Appendix for. Institutions and Behavior: Experimental Evidence on the Effects of Democracy Appendx for Insttutons and Behavor: Expermental Evdence on the Effects of Democrac 1. Instructons 1.1 Orgnal sessons Welcome You are about to partcpate n a stud on decson-makng, and ou wll be pad for our

More information

Comparative Analysis of Feature Extraction Methods for Optic Disc Detection

Comparative Analysis of Feature Extraction Methods for Optic Disc Detection IOSR Journal of Computer Engneerng (IOSR-JCE) e-issn: 2278-0661, p- ISSN: 2278-8727Volume 16, Issue 3, Ver. III (May-Jun. 2014), PP 49-54 Comparatve Analyss of Feature Extracton Methods for Optc Dsc Detecton

More information

A Novel artifact for evaluating accuracies of gear profile and pitch measurements of gear measuring instruments

A Novel artifact for evaluating accuracies of gear profile and pitch measurements of gear measuring instruments A Novel artfact for evaluatng accuraces of gear profle and ptch measurements of gear measurng nstruments Sonko Osawa, Osamu Sato, Yohan Kondo, Toshyuk Takatsuj (NMIJ/AIST) Masaharu Komor (Kyoto Unversty)

More information

AN ENHANCED GAGS BASED MTSVSL LEARNING TECHNIQUE FOR CANCER MOLECULAR PATTERN PREDICTION OF CANCER CLASSIFICATION

AN ENHANCED GAGS BASED MTSVSL LEARNING TECHNIQUE FOR CANCER MOLECULAR PATTERN PREDICTION OF CANCER CLASSIFICATION www.arpapress.com/volumes/vol8issue2/ijrras_8_2_02.pdf AN ENHANCED GAGS BASED MTSVSL LEARNING TECHNIQUE FOR CANCER MOLECULAR PATTERN PREDICTION OF CANCER CLASSIFICATION I. Jule 1 & E. Krubakaran 2 1 Department

More information

Semantics and image content integration for pulmonary nodule interpretation in thoracic computed tomography

Semantics and image content integration for pulmonary nodule interpretation in thoracic computed tomography Semantcs and mage content ntegraton for pulmonary nodule nterpretaton n thoracc computed tomography Danela S. Racu a, Ekarn Varutbangkul a, Jane G. Csneros a, Jacob D. Furst a, Davd S. Channn b, Samuel

More information

4.2 Scheduling to Minimize Maximum Lateness

4.2 Scheduling to Minimize Maximum Lateness 4. Schedulng to Mnmze Maxmum Lateness Schedulng to Mnmzng Maxmum Lateness Mnmzng lateness problem. Sngle resource processes one ob at a tme. Job requres t unts of processng tme and s due at tme d. If starts

More information

A MIXTURE OF EXPERTS FOR CATARACT DIAGNOSIS IN HOSPITAL SCREENING DATA

A MIXTURE OF EXPERTS FOR CATARACT DIAGNOSIS IN HOSPITAL SCREENING DATA Journal of Theoretcal and Appled Informaton Technology 2005 ongong JATIT & LLS ISSN: 1992-8645 www.jatt.org E-ISSN: 1817-3195 A MIXTURE OF EXPERTS FOR CATARACT DIAGNOSIS IN HOSPITAL SCREENING DATA 1 SUNGMIN

More information

Research Article Statistical Segmentation of Regions of Interest on a Mammographic Image

Research Article Statistical Segmentation of Regions of Interest on a Mammographic Image Hndaw Publshng Corporaton EURASIP Journal on Advances n Sgnal Processng Volume 2007, Artcle ID 49482, 8 pages do:10.1155/2007/49482 Research Artcle Statstcal Segmentaton of Regons of Interest on a Mammographc

More information

ARTICLE IN PRESS. computer methods and programs in biomedicine xxx (2007) xxx xxx. journal homepage:

ARTICLE IN PRESS. computer methods and programs in biomedicine xxx (2007) xxx xxx. journal homepage: computer methods and programs n bomedcne xxx (2007) xxx xxx journal homepage: www.ntl.elseverhealth.com/journals/cmpb Improvng bran tumor characterzaton on MRI by probablstc neural networks and non-lnear

More information

An expressive three-mode principal components model for gender recognition

An expressive three-mode principal components model for gender recognition Journal of Vson (4) 4, 36-377 http://journalofvson.org/4/5// 36 An expressve three-mode prncpal components model for gender recognton James W. Davs Hu Gao Department of Computer and Informaton Scence,

More information

An Improved Time Domain Pitch Detection Algorithm for Pathological Voice

An Improved Time Domain Pitch Detection Algorithm for Pathological Voice Amercan Journal of Appled Scences 9 (1): 93-102, 2012 ISSN 1546-9239 2012 Scence Publcatons An Improved Tme Doman Ptch Detecton Algorthm for Pathologcal Voce Mohd Redzuan Jamaludn, Shekh Hussan Shakh Salleh,

More information

Lateral Transfer Data Report. Principal Investigator: Andrea Baptiste, MA, OT, CIE Co-Investigator: Kay Steadman, MA, OTR, CHSP. Executive Summary:

Lateral Transfer Data Report. Principal Investigator: Andrea Baptiste, MA, OT, CIE Co-Investigator: Kay Steadman, MA, OTR, CHSP. Executive Summary: Samar tmed c ali ndus t r esi nc 55Fl em ngdr ve, Un t#9 Cambr dge, ON. N1T2A9 T el. 18886582206 Ema l. nf o@s amar t r ol l boar d. c om www. s amar t r ol l boar d. c om Lateral Transfer Data Report

More information

IMPROVING THE EFFICIENCY OF BIOMARKER IDENTIFICATION USING BIOLOGICAL KNOWLEDGE

IMPROVING THE EFFICIENCY OF BIOMARKER IDENTIFICATION USING BIOLOGICAL KNOWLEDGE IMPROVING THE EFFICIENCY OF BIOMARKER IDENTIFICATION USING BIOLOGICAL KNOWLEDGE JOHN H. PHAN The Wallace H. Coulter Department of Bomedcal Engneerng, Georga Insttute of Technology, 313 Ferst Drve Atlanta,

More information

Subject-Adaptive Real-Time Sleep Stage Classification Based on Conditional Random Field

Subject-Adaptive Real-Time Sleep Stage Classification Based on Conditional Random Field Subject-Adaptve Real-Tme Sleep Stage Classfcaton Based on Condtonal Random Feld Gang Luo, PhD, Wanl Mn, PhD IBM TJ Watson Research Center, Hawthorne, NY {luog, wanlmn}@usbmcom Abstract Sleep stagng s the

More information

Combined Temporal and Spatial Filter Structures for CDMA Systems

Combined Temporal and Spatial Filter Structures for CDMA Systems Combned Temporal and Spatal Flter Structures for CDMA Systems Ayln Yener WINLAB, Rutgers Unversty yener@wnlab.rutgers.edu Roy D. Yates WINLAB, Rutgers Unversty ryates@wnlab.rutgers.edu Sennur Ulukus AT&T

More information

Survival Rate of Patients of Ovarian Cancer: Rough Set Approach

Survival Rate of Patients of Ovarian Cancer: Rough Set Approach Internatonal OEN ACCESS Journal Of Modern Engneerng esearch (IJME) Survval ate of atents of Ovaran Cancer: ough Set Approach Kamn Agrawal 1, ragat Jan 1 Department of Appled Mathematcs, IET, Indore, Inda

More information

DETECTION AND CLASSIFICATION OF BRAIN TUMOR USING ML

DETECTION AND CLASSIFICATION OF BRAIN TUMOR USING ML DOI: http://dx.do.org/0.26483/arcs.v92.5807 Volume 9, No. 2, March-Aprl 208 Internatonal Journal of Advanced Research n Computer Scence RESEARCH PAPER Avalable Onlne at www.arcs.nfo ISSN No. 0976-5697

More information

Balanced Query Methods for Improving OCR-Based Retrieval

Balanced Query Methods for Improving OCR-Based Retrieval Balanced Query Methods for Improvng OCR-Based Retreval Kareem Darwsh Electrcal and Computer Engneerng Dept. Unversty of Maryland, College Park College Park, MD 20742 kareem@glue.umd.edu Douglas W. Oard

More information

Richard Williams Notre Dame Sociology Meetings of the European Survey Research Association Ljubljana,

Richard Williams Notre Dame Sociology   Meetings of the European Survey Research Association Ljubljana, Rchard Wllams Notre Dame Socology rwllam@nd.edu http://www.nd.edu/~rwllam Meetngs of the European Survey Research Assocaton Ljubljana, Slovena July 19, 2013 Comparng Logt and Probt Coeffcents across groups

More information

Estimation for Pavement Performance Curve based on Kyoto Model : A Case Study for Highway in the State of Sao Paulo

Estimation for Pavement Performance Curve based on Kyoto Model : A Case Study for Highway in the State of Sao Paulo Estmaton for Pavement Performance Curve based on Kyoto Model : A Case Study for Kazuya AOKI, PASCO CORPORATION, Yokohama, JAPAN, Emal : kakzo603@pasco.co.jp Octávo de Souza Campos, Publc Servces Regulatory

More information

Recognition of ASL for Human-robot Interaction

Recognition of ASL for Human-robot Interaction 66 IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.17 No.7, July 2017 Recognton of ASL for Human-robot Interacton Md. Al-Amn Bhuyan College of Computer Scences & Informaton Technology,

More information

A Linear Regression Model to Detect User Emotion for Touch Input Interactive Systems

A Linear Regression Model to Detect User Emotion for Touch Input Interactive Systems 2015 Internatonal Conference on Affectve Computng and Intellgent Interacton (ACII) A Lnear Regresson Model to Detect User Emoton for Touch Input Interactve Systems Samt Bhattacharya Dept of Computer Scence

More information

Lymphoma Cancer Classification Using Genetic Programming with SNR Features

Lymphoma Cancer Classification Using Genetic Programming with SNR Features Lymphoma Cancer Classfcaton Usng Genetc Programmng wth SNR Features Jn-Hyuk Hong and Sung-Bae Cho Dept. of Computer Scence, Yonse Unversty, 134 Shnchon-dong, Sudaemoon-ku, Seoul 120-749, Korea hjnh@candy.yonse.ac.kr,

More information

Classification of Breast Tumor in Mammogram Images Using Unsupervised Feature Learning

Classification of Breast Tumor in Mammogram Images Using Unsupervised Feature Learning Amercan Journal of Appled Scences Orgnal Research Paper Classfcaton of Breast Tumor n Mammogram Images Usng Unsupervsed Feature Learnng 1 Adarus M. Ibrahm, 1 Baharum Baharudn, 1 Abas Md Sad and 2 P.N.

More information

Journal of Engineering Science and Technology Review 11 (2) (2018) Research Article

Journal of Engineering Science and Technology Review 11 (2) (2018) Research Article Jestr Journal of Engneerng Scence and Technology Revew () (08) 5 - Research Artcle Prognoss Evaluaton of Ovaran Granulosa Cell Tumor Based on Co-forest ntellgence Model Xn Lao Xn Zheng Juan Zou Mn Feng

More information

Automated and ERP-Based Diagnosis of Attention-Deficit Hyperactivity Disorder in Children

Automated and ERP-Based Diagnosis of Attention-Deficit Hyperactivity Disorder in Children Orgnal Artcle Automated and ERP-Based Dagnoss of Attenton-Defct Hyperactvty Dsorder n Chldren Abstract Event-related potental (ERP) s one of the most nformatve and dynamc methods of montorng cogntve processes,

More information

International Journal of Emerging Technologies in Computational and Applied Sciences (IJETCAS)

International Journal of Emerging Technologies in Computational and Applied Sciences (IJETCAS) Internatonal Assocaton of Scentfc Innovaton and Research (IASIR (An Assocaton Unfyng the Scences, Engneerng, and Appled Research Internatonal Journal of Emergng Technologes n Computatonal and Appled Scences

More information

Detection of Lung Cancer at Early Stage using Neural Network Techniques for Preventing Health Care

Detection of Lung Cancer at Early Stage using Neural Network Techniques for Preventing Health Care IJSRD - Internatonal Journal for Scentfc Research & Development Vol. 3, Issue 4, 15 ISSN (onlne): 31-613 Detecton of Lung Cancer at Early Stage usng Neural Network echnques for Preventng Health Care Megha

More information

Gene Selection Based on Mutual Information for the Classification of Multi-class Cancer

Gene Selection Based on Mutual Information for the Classification of Multi-class Cancer Gene Selecton Based on Mutual Informaton for the Classfcaton of Mult-class Cancer Sheng-Bo Guo,, Mchael R. Lyu 3, and Tat-Mng Lok 4 Department of Automaton, Unversty of Scence and Technology of Chna, Hefe,

More information

Modeling Multi Layer Feed-forward Neural. Network Model on the Influence of Hypertension. and Diabetes Mellitus on Family History of

Modeling Multi Layer Feed-forward Neural. Network Model on the Influence of Hypertension. and Diabetes Mellitus on Family History of Appled Mathematcal Scences, Vol. 7, 2013, no. 41, 2047-2053 HIKARI Ltd, www.m-hkar.com Modelng Mult Layer Feed-forward Neural Network Model on the Influence of Hypertenson and Dabetes Melltus on Famly

More information

Pattern Recognition for Robotic Fish Swimming Gaits Based on Artificial Lateral Line System and Subtractive Clustering Algorithms

Pattern Recognition for Robotic Fish Swimming Gaits Based on Artificial Lateral Line System and Subtractive Clustering Algorithms Sensors & Transducers, Vol. 18, Issue 11, November 14, pp. 7-16 Sensors & Transducers 14 by IFSA Publshng, S. L. http://www.sensorsportal.com Pattern Recognton for Robotc Fsh Swmmng Gats Based on Artfcal

More information

Analysis of the QRS Complex for Apnea-Bradycardia Characterization in Preterm Infants

Analysis of the QRS Complex for Apnea-Bradycardia Characterization in Preterm Infants Author manuscrpt, publshed n "Conference proceedngs : Annual Internatonal Conference of the IEEE Engneerng n Medcne and Bology Socety. 2009;:946-9" DOI : 0.09/IEMBS.2009.533353 Analyss of the QRS Complex

More information

1 INTRODUCTION 2 HEART-BEAT CYCLE DETECTION

1 INTRODUCTION 2 HEART-BEAT CYCLE DETECTION Abstract: Ths artcle ntrouces a metho for constructng a heart-beat sgnal from a long recor of phonocarogram (PCG). The metho comprses a seres of algorthms that ecompose a long ata recor of heart soun to

More information

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and Ths artcle appeared n a journal publshed by Elsever. The attached copy s furnshed to the author for nternal non-commercal research and educaton use, ncludng for nstructon at the authors nsttuton and sharng

More information

A-UNIFAC Modeling of Binary and Multicomponent Phase Equilibria of Fatty Esters+Water+Methanol+Glycerol

A-UNIFAC Modeling of Binary and Multicomponent Phase Equilibria of Fatty Esters+Water+Methanol+Glycerol -UNIFC Modelng of Bnary and Multcomponent Phase Equlbra of Fatty Esters+Water+Methanol+Glycerol N. Garrdo a, O. Ferrera b, R. Lugo c, J.-C. de Hemptnne c, M. E. Macedo a, S.B. Bottn d,* a Department of

More information

Automatic Labelling and BI-RADS Characterisation of Mammogram Densities

Automatic Labelling and BI-RADS Characterisation of Mammogram Densities Lmted crculaton. For revew only. Automatc Labellng and BI-RADS Charactersaton of ammogram Denstes K. aras,. G. Lnguraru 3,. G. Ballester 4, S. Petroud,. Tsknaks and Sr. Brady Insttute of Computer Scence,

More information

Concentration of teicoplanin in the serum of adults with end stage chronic renal failure undergoing treatment for infection

Concentration of teicoplanin in the serum of adults with end stage chronic renal failure undergoing treatment for infection Journal of Antmcrobal Chemotherapy (1996) 37, 117-121 Concentraton of tecoplann n the serum of adults wth end stage chronc renal falure undergong treatment for nfecton A. MercateUo'*, K. Jaber*, D. Hfflare-Buys*,

More information

An Approach to Discover Dependencies between Service Operations*

An Approach to Discover Dependencies between Service Operations* 36 JOURNAL OF SOFTWARE VOL. 3 NO. 9 DECEMBER 2008 An Approach to Dscover Dependences between Servce Operatons* Shuyng Yan Research Center for Grd and Servce Computng Insttute of Computng Technology Chnese

More information

Myocardial Motion Analysis of Echocardiography Images using Optical Flow Radial Direction Distribution

Myocardial Motion Analysis of Echocardiography Images using Optical Flow Radial Direction Distribution Journal of Computer Scence 7 (7): 1046-1051, 011 ISSN 1549-3636 011 Scence Publcatons Myocardal Moton Analyss of Echocardography Images usng Optcal Flow Radal Drecton Dstrbuton Slamet Ryad, Mohd Marzuk

More information

Myocardial Mural Thickness During the Cardiac Cycle

Myocardial Mural Thickness During the Cardiac Cycle Myocardal Mural Thckness Durng the Cardac Cycle By Erc O. Fegl, M.D., and Donald L. Fry, M.D. An understandng of the relatonshp between forces and veloctes of contracton n muscle fbers to the pressures

More information

Introduction ORIGINAL RESEARCH

Introduction ORIGINAL RESEARCH ORIGINAL RESEARCH Assessng the Statstcal Sgnfcance of the Acheved Classfcaton Error of Classfers Constructed usng Serum Peptde Profles, and a Prescrpton for Random Samplng Repeated Studes for Massve Hgh-Throughput

More information

A deterministic approach for finding the T onset parameter of Flatten T wave in ECG

A deterministic approach for finding the T onset parameter of Flatten T wave in ECG A determnstc approach for fndng the T onset parameter of Flatten T wave n ECG Uzar Iqbal a, Teh Yng Wah a, *, Muhammad Habb ur Rehman a, Quratulan Masto a a Department of Informaton Systems, Faculty of

More information

ARTICLE IN PRESS Neuropsychologia xxx (2010) xxx xxx

ARTICLE IN PRESS Neuropsychologia xxx (2010) xxx xxx Neuropsychologa xxx (200) xxx xxx Contents lsts avalable at ScenceDrect Neuropsychologa journal homepage: www.elsever.com/locate/neuropsychologa Storage and bndng of object features n vsual workng memory

More information

JOINT SUB-CLASSIFIERS ONE CLASS CLASSIFICATION MODEL FOR AVIAN INFLUENZA OUTBREAK DETECTION

JOINT SUB-CLASSIFIERS ONE CLASS CLASSIFICATION MODEL FOR AVIAN INFLUENZA OUTBREAK DETECTION JOINT SUB-CLASSIFIERS ONE CLASS CLASSIFICATION MODEL FOR AVIAN INFLUENZA OUTBREAK DETECTION Je Zhang, Je Lu, Guangquan Zhang Centre for Quantum Computaton & Intellgent Systems Faculty of Engneerng and

More information

Prognosis and Diagnosis of Breast Cancer Using Interactive Dashboard Through Big Data Analytics

Prognosis and Diagnosis of Breast Cancer Using Interactive Dashboard Through Big Data Analytics Prognoss and Dagnoss of Breast Cancer Usng Interactve Dashboard Through Bg Data Analytcs Gomath N, and Sandhya P 2 * Department of Computer Scence and Engneerng, Veltech Dr. RR & Dr. SR Unversty, Avad,

More information

Investigation of zinc oxide thin film by spectroscopic ellipsometry

Investigation of zinc oxide thin film by spectroscopic ellipsometry VNU Journal of Scence, Mathematcs - Physcs 24 (2008) 16-23 Investgaton of znc oxde thn flm by spectroscopc ellpsometry Nguyen Nang Dnh 1, Tran Quang Trung 2, Le Khac Bnh 2, Nguyen Dang Khoa 2, Vo Th Ma

More information

Algorithms 2009, 2, ; doi: /a OPEN ACCESS

Algorithms 2009, 2, ; doi: /a OPEN ACCESS Algorthms 009,, 350-367; do:0.3390/a04350 OPEN ACCESS algorthms ISSN 999-4893 www.mdp.com/journal/algorthms Artcle CADrx for GBM Bran Tumors: Predctng Treatment Response from Changes n Dffuson-Weghted

More information

Comparison among Feature Encoding Techniques for HIV-1 Protease Cleavage Specificity

Comparison among Feature Encoding Techniques for HIV-1 Protease Cleavage Specificity Internatonal Journal of Intellgent Systems and Applcatons n Engneerng Advanced Technology and Scence ISSN:2147-67992147-6799 http://jsae.atscence.org/ Orgnal Research Paper Comparson among Feature Encodng

More information

Sparse Representation of HCP Grayordinate Data Reveals. Novel Functional Architecture of Cerebral Cortex

Sparse Representation of HCP Grayordinate Data Reveals. Novel Functional Architecture of Cerebral Cortex 1 Sparse Representaton of HCP Grayordnate Data Reveals Novel Functonal Archtecture of Cerebral Cortex X Jang 1, Xang L 1, Jngle Lv 2,1, Tuo Zhang 2,1, Shu Zhang 1, Le Guo 2, Tanmng Lu 1* 1 Cortcal Archtecture

More information

Integration of sensory information within touch and across modalities

Integration of sensory information within touch and across modalities Integraton of sensory nformaton wthn touch and across modaltes Marc O. Ernst, Jean-Perre Brescan, Knut Drewng & Henrch H. Bülthoff Max Planck Insttute for Bologcal Cybernetcs 72076 Tübngen, Germany marc.ernst@tuebngen.mpg.de

More information

Arrhythmia Detection based on Morphological and Time-frequency Features of T-wave in Electrocardiogram ABSTRACT

Arrhythmia Detection based on Morphological and Time-frequency Features of T-wave in Electrocardiogram ABSTRACT Orgnal Artcle Arrhythma Detecton based on Morphologcal and Tme-frequency Features of T-wave n Electrocardogram Elham Zeraatkar, Saeed Kerman, Alreza Mehrdehnav 1, A. Amnzadeh 2, E. Zeraatkar 3, Hamd Sane

More information

Research Article Segmentation of Bone with Region Based Active Contour Model in PD Weighted MR Images of Shoulder

Research Article Segmentation of Bone with Region Based Active Contour Model in PD Weighted MR Images of Shoulder Computatonal and Mathematcal Methods n Medcne Volume 2015, Artcle ID 754894, 13 pages http://dx.do.org/10.1155/2015/754894 Research Artcle Segmentaton of Bone wth Regon Based Actve Contour Model n PD Weghted

More information

An Introduction to Modern Measurement Theory

An Introduction to Modern Measurement Theory An Introducton to Modern Measurement Theory Ths tutoral was wrtten as an ntroducton to the bascs of tem response theory (IRT) modelng and ts applcatons to health outcomes measurement for the Natonal Cancer

More information

Chapter 20. Aggregation and calibration. Betina Dimaranan, Thomas Hertel, Robert McDougall

Chapter 20. Aggregation and calibration. Betina Dimaranan, Thomas Hertel, Robert McDougall Chapter 20 Aggregaton and calbraton Betna Dmaranan, Thomas Hertel, Robert McDougall In the prevous chapter we dscussed how the fnal verson 3 GTAP data base was assembled. Ths data base s extremely large.

More information

DS May 31,2012 Commissioner, Development. Services Department SPA June 7,2012

DS May 31,2012 Commissioner, Development. Services Department SPA June 7,2012 . h,oshawa o Report To: From: Subject: Development Servces Commttee Item: Date of Report: DS-12-189 May 31,2012 Commssoner, Development Fle: Date of Meetng: Servces Department SPA-2010-09 June 7,2012 Applcaton

More information

Heart Rate Variability Analysis Diagnosing Atrial Fibrillation

Heart Rate Variability Analysis Diagnosing Atrial Fibrillation X-ray PIV Measurements of Velocty Feld of Blood Flows Volume 5, umber 2: 46-52, October 2007 Internatonal Journal of Vascular Bomedcal Engneerng Heart Rate Varablty Analyss Dagnosng Atral Fbrllaton Jnho

More information

A Neural Network System for Diagnosis and Assessment of Tremor in Parkinson Disease Patients

A Neural Network System for Diagnosis and Assessment of Tremor in Parkinson Disease Patients A Neural Network System for Dagnoss and Assessment of Tremor n Parknson Dsease Patents Omd Bazgr*, Javad Frounch Department of Electrcty and Computer Engneerng Unversty of Tabrz Tabrz, Iran Omdbazgr92@ms.tabrzu.ac.r

More information

NUMERICAL COMPARISONS OF BIOASSAY METHODS IN ESTIMATING LC50 TIANHONG ZHOU

NUMERICAL COMPARISONS OF BIOASSAY METHODS IN ESTIMATING LC50 TIANHONG ZHOU NUMERICAL COMPARISONS OF BIOASSAY METHODS IN ESTIMATING LC50 by TIANHONG ZHOU B.S., Chna Agrcultural Unversty, 2003 M.S., Chna Agrcultural Unversty, 2006 A THESIS submtted n partal fulfllment of the requrements

More information

Statistically Weighted Voting Analysis of Microarrays for Molecular Pattern Selection and Discovery Cancer Genotypes

Statistically Weighted Voting Analysis of Microarrays for Molecular Pattern Selection and Discovery Cancer Genotypes IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.6 No.2, December 26 73 Statstcally Weghted Votng Analyss of Mcroarrays for Molecular Pattern Selecton and Dscovery Cancer Genotypes

More information

INTEGRATIVE NETWORK ANALYSIS TO IDENTIFY ABERRANT PATHWAY NETWORKS IN OVARIAN CANCER

INTEGRATIVE NETWORK ANALYSIS TO IDENTIFY ABERRANT PATHWAY NETWORKS IN OVARIAN CANCER INTEGRATIVE NETWORK ANALYSIS TO IDENTIFY ABERRANT PATHWAY NETWORKS IN OVARIAN CANCER LI CHEN 1,2, JIANHUA XUAN 1,*, JINGHUA GU 1, YUE WANG 1, ZHEN ZHANG 2, TIAN LI WANG 2, IE MING SHIH 2 1The Bradley Department

More information

ENRICHING PROCESS OF ICE-CREAM RECOMMENDATION USING COMBINATORIAL RANKING OF AHP AND MONTE CARLO AHP

ENRICHING PROCESS OF ICE-CREAM RECOMMENDATION USING COMBINATORIAL RANKING OF AHP AND MONTE CARLO AHP ENRICHING PROCESS OF ICE-CREAM RECOMMENDATION USING COMBINATORIAL RANKING OF AHP AND MONTE CARLO AHP 1 AKASH RAMESHWAR LADDHA, 2 RAHUL RAGHVENDRA JOSHI, 3 Dr.PEETI MULAY 1 M.Tech, Department of Computer

More information

ARTICLE IN PRESS Biomedical Signal Processing and Control xxx (2011) xxx xxx

ARTICLE IN PRESS Biomedical Signal Processing and Control xxx (2011) xxx xxx Bomedcal Sgnal Processng and Control xxx (2011) xxx xxx Contents lsts avalable at ScenceDrect Bomedcal Sgnal Processng and Control journa l h omepage: www.elsever.com/locate/bspc Dscovery of multple level

More information

Muscle Synergy Analysis Between Young and Elderly People in Standing-Up Motion

Muscle Synergy Analysis Between Young and Elderly People in Standing-Up Motion An, Q., Ikemoto, Y., and Asama, H. Paper: Muscle Synergy Analyss Between Young and Elderly People n Standng-Up Moton Q An, Yusuke Ikemoto, and Hajme Asama The Unversty of Tokyo 7-3-1 Hongo, Bunkyou-ku,

More information

Machine Understanding - a new area of research aimed at building thinking/understanding machines

Machine Understanding - a new area of research aimed at building thinking/understanding machines achne Understandng - a new area of research amed at buldng thnkng/understandng machnes Zbgnew Les and agdalena Les St. Queen Jadwga Research Insttute of Understandng, elbourne, Australa sqru@outlook.com

More information

Detection of Cancer Metastasis Using a Novel Macroscopic Hyperspectral Method

Detection of Cancer Metastasis Using a Novel Macroscopic Hyperspectral Method Detecton of Cancer Metastass Usng a Novel Macroscopc Hyperspectral Method Hamed Akbar a, Luma V. Halg a, Hongzheng Zhang b, Dongsheng Wang b, Zhuo Georga Chen b, Baowe Fe a,c,d,* a Department of Radology

More information

A Support Vector Machine Classifier based on Recursive Feature Elimination for Microarray Data in Breast Cancer Characterization. Abstract.

A Support Vector Machine Classifier based on Recursive Feature Elimination for Microarray Data in Breast Cancer Characterization. Abstract. A Support Vector Machne Classfer based on Recursve Feature Elmnaton for Mcroarray Data n Breast Cancer Characterzaton. R.Campann, D. Dongovann, E. Iamper, N. Lanconell, G. Palermo, M. Roffll, A. Rccard

More information

Joint Modelling Approaches in diabetes research. Francisco Gude Clinical Epidemiology Unit, Hospital Clínico Universitario de Santiago

Joint Modelling Approaches in diabetes research. Francisco Gude Clinical Epidemiology Unit, Hospital Clínico Universitario de Santiago Jont Modellng Approaches n dabetes research Clncal Epdemology Unt, Hosptal Clínco Unverstaro de Santago Outlne 1 Dabetes 2 Our research 3 Some applcatons Dabetes melltus Is a serous lfe-long health condton

More information

Resampling Methods for the Area Under the ROC Curve

Resampling Methods for the Area Under the ROC Curve Resamplng ethods for the Area Under the ROC Curve Andry I. Bandos AB6@PITT.EDU Howard E. Rockette HERBST@PITT.EDU Department of Bostatstcs, Graduate School of Publc Health, Unversty of Pttsburgh, Pttsburgh,

More information

THE NORMAL DISTRIBUTION AND Z-SCORES COMMON CORE ALGEBRA II

THE NORMAL DISTRIBUTION AND Z-SCORES COMMON CORE ALGEBRA II Name: Date: THE NORMAL DISTRIBUTION AND Z-SCORES COMMON CORE ALGEBRA II The normal dstrbuton can be used n ncrements other than half-standard devatons. In fact, we can use ether our calculators or tables

More information

Nonlinear Modeling Method Based on RBF Neural Network Trained by AFSA with Adaptive Adjustment

Nonlinear Modeling Method Based on RBF Neural Network Trained by AFSA with Adaptive Adjustment Advances n Engneerng Research (AER), volue 48 3rd Workshop on Advanced Research and Technology n Industry Applcatons (WARTIA 27) Nonlnear Modelng Method Based on RBF Neural Network Traned by AFSA wth Adaptve

More information