#### DMCA

## An Unsupervised Ensemble-based Markov Random Field Approach to Microscope Cell Image Segmentation

### Citations

5141 | Optimization by simulated annealing
- Kirkpatrick, Gellat, et al.
- 1983
(Show Context)
Citation Context ... U is the following: U = ∑ j=0 n ( Esingleton (i j)+Edoubleton (i j) ) . The optimization of MRF configuration can be done by optimizing U . If P(ω) follows a Gibbs distribution, simulated annealing (=-=Kirkpatrick et al., 1983-=-) converges to the optimal solution with 1 probability. However, simulated annealing tends to be slow in some cases. However, Iterated Conditional Modes (ICM) (Besag, 1986) can also be effective if th... |

2244 |
A Threshold Selection Method from Gray-Level Histogram
- Otsu
- 1979
(Show Context)
Citation Context ...cs curve for the proposed approach. Approach RI proposed 0.96 Mean Threshold 0.96 Merging Algorithm (Lin et al., 2003) 0.96 AS Manual 0.96 RC Threshold (Ridler and Calvard, 1978) 0.92 Otsu Threshold (=-=Otsu, 1979-=-) 0.92 Watershed (direct) 0.91 Watershed (gradient) 0.90 Active Masks (Srinivasa et al., 2008) 0.87 Table 2: Comparison of the proposed method with other approaches. 6 Conclusion In this paper, we pre... |

1246 | On the statistical analysis of dirty pictures
- Besag
- 1986
(Show Context)
Citation Context ...ated annealing (Kirkpatrick et al., 1983) converges to the optimal solution with 1 probability. However, simulated annealing tends to be slow in some cases. However, Iterated Conditional Modes (ICM) (=-=Besag, 1986-=-) can also be effective if there is a good initial configuration. 3 Unsupervised MRF-ensembles As we stated in Section 2, the usual optimization of MRFs needs training. In this section, we present an ... |

808 |
Objective criteria for the evaluation of clustering methods
- Rand
- 1971
(Show Context)
Citation Context ...ty map for the sample image shown in 2(a). (b)-(i) Voted images with ascending confidence level. all samples marked as object points: FSCORE = 2 ·SEN ·PPV SEN +PPV . 4.2.6 Rand Index Rand Index (RI) (=-=Rand, 1971-=-) measure the agreement between the segmentation and the ground truth in the following way: FSCORE = n11 +n00 n11 +n00 +n01 +n10 . 4.2.7 Receiver Operating Characteristics We also disclose the Receive... |

469 |
Combining Pattern Classifiers: Methods and Algorithms.
- Kuncheva
- 2004
(Show Context)
Citation Context ...s which belong to either the segmentation or the ground truth bot not both. We also normalize SD with the number of pixels in the image. That is SD = n01 +n10 n . 4.2.2 Sensitivity Sensitivity (SEN) (=-=Kuncheva, 2004-=-) is a statistical measure for quantifying the correctly identified positive samples. In our case, it is defined as follows: SEN = n11 n11 +n01 . 4.2.3 Specificity Specificity (SPE) (Kuncheva, 2004) m... |

175 |
Picture thresholding using an iterative selection method.
- Ridler, Calvard
- 1978
(Show Context)
Citation Context ...citysFigure 4: Receiver Operating Characteristics curve for the proposed approach. Approach RI proposed 0.96 Mean Threshold 0.96 Merging Algorithm (Lin et al., 2003) 0.96 AS Manual 0.96 RC Threshold (=-=Ridler and Calvard, 1978-=-) 0.92 Otsu Threshold (Otsu, 1979) 0.92 Watershed (direct) 0.91 Watershed (gradient) 0.90 Active Masks (Srinivasa et al., 2008) 0.87 Table 2: Comparison of the proposed method with other approaches. 6... |

120 |
Markov fields on finite graphs and lattices
- Hammersley, Clifford
- 1971
(Show Context)
Citation Context ... ik ∈ Ni j ) , where Ni j is a neighbourhood of i j. The segmentation of an image I with the MRF framework presented above, one must find an optimal labelling. Due to the Hammersley-Clifford Theorem (=-=Hammersley and Clifford, 1971-=-), we can calculate the global energy for a labelling by summarizing the local energies for each pixels if P(ω) follows a Gibbs distribution. We split the local energy into two terms for all i j: Esin... |

49 | Bayesian image classification using Markov random fields,”
- Berthod, Kato, et al.
- 1996
(Show Context)
Citation Context ...esults in section 5. Finally, we draw conclusions in section 6. 2 Markov Random Field Segmentation In this section, we briefly summarize the basis for Markov Random Field (MRF) segmentation based on (=-=Berthod et al., 1996-=-). Let I = {i1, i2, . . . , in} be an image. Let Λ = {0,1} be a set of labels. Then, we assign each i j, j = 1, . . . , n a label ωi j . Let X be a labelling field. X is a Markov Random Field if P(X =... |

40 |
A hybrid 3D watershed algorithm incorporating gradient cues and object models for automatic segmentation of nuclei in confocal image stacks,”
- Lin, Adiga, et al.
- 2003
(Show Context)
Citation Context ...4 0,5 0,6 0,7 0,8 0,9 1 Se n si ti vi ty 1 - specificitysFigure 4: Receiver Operating Characteristics curve for the proposed approach. Approach RI proposed 0.96 Mean Threshold 0.96 Merging Algorithm (=-=Lin et al., 2003-=-) 0.96 AS Manual 0.96 RC Threshold (Ridler and Calvard, 1978) 0.92 Otsu Threshold (Otsu, 1979) 0.92 Watershed (direct) 0.91 Watershed (gradient) 0.90 Active Masks (Srinivasa et al., 2008) 0.87 Table 2... |

24 |
Nuclear segmentation in microscope cell images: a hand-segmented dataset and comparison of algorithms
- Coelho, Shariff, et al.
(Show Context)
Citation Context ...as physicians. Cell segmentation is essential for several cytometric tasks like cell counting and tracking. The automatic segmentation of cell images is a well-studied field (Meijering et al., 2012) (=-=Coelho et al., 2009-=-). However, efficient segmentation of such images is still an open issue. A sample image can be seen in Figure 1. Figure 1: A sample image from the dataset. In this paper, we present an approach which... |

18 |
An ensemble-based system for microaneurysm detection and diabetic retinopathy grading
- Antal, Hajdu
- 2012
(Show Context)
Citation Context ... and start the optimization process from an initial configuration. As no single plane can be selected obviously as a proper initial labelling for an MRF, we propose to use all of them as an ensemble (=-=Antal and Hajdu, 2012-=-a) (Antal and Hajdu, 2012b). That is, we run the optimization eight times using each plane as the initial labelling. Then, we can use pixelwise voting (Nagy et al., 2011) on the resulting eight images... |

8 | ROC analysis: web-based calculator for ROC curves. [http://www. jrocfit.org - Eng |

7 | Improving microaneurysm detection using an optimally selected subset of candidate extractors and preprocessing methods, - Antel, Hajdu - 2012 |

6 | Active mask segmentation for the cell-volume computation and Golgi-body segmentation of HeLa cell images
- Srinivasa, Fickus, et al.
- 2008
(Show Context)
Citation Context ... Merging Algorithm (Lin et al., 2003) 0.96 AS Manual 0.96 RC Threshold (Ridler and Calvard, 1978) 0.92 Otsu Threshold (Otsu, 1979) 0.92 Watershed (direct) 0.91 Watershed (gradient) 0.90 Active Masks (=-=Srinivasa et al., 2008-=-) 0.87 Table 2: Comparison of the proposed method with other approaches. 6 Conclusion In this paper, we presented an approach to the unsupervised segmentation of images using Markov Random Field. In t... |

5 |
Advantages to transforming the receiver operating characteristic (ROC) curve into likelihood ratio co‐ordinates
- Johnson
- 2004
(Show Context)
Citation Context ... segmentation and the ground truth in the following way: FSCORE = n11 +n00 n11 +n00 +n01 +n10 . 4.2.7 Receiver Operating Characteristics We also disclose the Receiver Operating Characteristics (ROC) (=-=Johnson, 2004-=-) curve for our segmentation approach. For the curve fitting and for the ROCrelated calculations, we used JROCFIT (Eng, ). 5 Results In Table 1, we can see the different evaluation metric values at th... |

3 |
Ensemble-based exudate detection in color fundus images
- Nagy, Harangi, et al.
- 2011
(Show Context)
Citation Context ...ll of them as an ensemble (Antal and Hajdu, 2012a) (Antal and Hajdu, 2012b). That is, we run the optimization eight times using each plane as the initial labelling. Then, we can use pixelwise voting (=-=Nagy et al., 2011-=-) on the resulting eight images. In this way, each pixel on the resulting image will be having a confidence level between 0 and 7 depending on how many of the segmentations labelled them as object poi... |

2 |
Chapter nine: Methods for Cell and Particle Tracking
- Meijering, Dzyubachyk, et al.
- 2012
(Show Context)
Citation Context ...essing community as well as physicians. Cell segmentation is essential for several cytometric tasks like cell counting and tracking. The automatic segmentation of cell images is a well-studied field (=-=Meijering et al., 2012-=-) (Coelho et al., 2009). However, efficient segmentation of such images is still an open issue. A sample image can be seen in Figure 1. Figure 1: A sample image from the dataset. In this paper, we pre... |