Abstract | ||
---|---|---|
In this paper, we introduce a new classifier ensemble approach, applied to tissue segmentation in optical images of the uterine cervix. Ensemble methods combine the predictions of a set of diverse classifiers. The main contribution of our approach is an effective way of combination based on each classifier's performance level-namely, the sensitivity p and specificity q, which also produces an optimal estimate of the true segmentation. In comparison with previous work [1] that utilizes the STAPLE algorithm [2] for performance level based combination, this work achieves multiple-observer segmentation in a Bayesian decision framework using the maximum a posterior (MAP) principle, considering each classifier as an observer. In our experiments, we applied our method and several other popular ensemble methods to the problem of detecting Acetowhite regions in cervical images. On 100 images, the overall performance of the proposed method is better than: (i) an overall classifier learned using the entire training set, (ii) average voting ensemble, (iii) ensemble based on the STAPLE algorithm; it is comparable to that of majority voting and that of the (manually picked) best-performing individual classifier in the ensemble set. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/ISBI.2009.5193054 | Boston, MA |
Keywords | Field | DocType |
sensitivity,bayesian decision framework,belief networks,ensemble method,diverse classifier,biomedical optical imaging,average voting ensemble,multiple classifier system,cervi- gram,maximum likelihood estimation,image segmentation,segmentation,new classifier ensemble approach,optical imaging,cervigram,ensemble set,acetowhite regions,index terms— classifier ensemble,tissue,image classification,multiple-observer segmentation,best-performing individual classifier,popular ensemble method,staple algorithm,overall classifier,biological tissues,performance level estimation,maximum a posterior principle,uterine cervix,biological organs,classifier ensemble,medical image processing,specificity,entire training set,optimal estimation,support vector machines,biomedical imaging,pixel,majority voting | Computer science,Image segmentation,Artificial intelligence,Classifier (linguistics),Contextual image classification,Ensemble learning,Computer vision,Pattern recognition,Segmentation,Support vector machine,Cascading classifiers,Margin classifier,Machine learning | Conference |
ISSN | ISBN | Citations |
1945-7928 E-ISBN : 978-1-4244-3932-4 | 978-1-4244-3932-4 | 3 |
PageRank | References | Authors |
0.46 | 11 | 8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wei Wang | 1 | 23 | 2.10 |
Yaoyao Zhu | 2 | 110 | 6.77 |
Xiaolei Huang | 3 | 1084 | 63.94 |
Daniel P. Lopresti | 4 | 1000 | 139.90 |
Zhiyun Xue | 5 | 245 | 22.97 |
L. Rodney Long | 6 | 534 | 56.98 |
Sameer Antani | 7 | 1402 | 134.03 |
George R. Thoma | 8 | 1207 | 132.81 |