Agenda
Overview
Technical Details
Our Team
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1804324/stermedia.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1809522/munichmiccai2015.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1794616/poziom-en.gif)
Overview
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793381/image04.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793383/image04_mask_truth.png)
Simple task
Nuclei segmentation
Overview
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793386/image01.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793390/image02.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793392/image03.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793393/image04.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793394/image05.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793395/image06.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793396/image07.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793397/image08.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793398/image09.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793399/image10.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793400/image11.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793401/image12.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793402/image13.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793403/image14.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793409/image01.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793410/image02.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793411/image03.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793412/image04.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793415/image05.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793416/image06.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793417/image07.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793424/image14.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793423/image13.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793422/image12.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793421/image11.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793420/image10.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793419/image09.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793418/image08.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793427/image15.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793428/image16.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793429/image17.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793430/image18.png)
Train
Test
Input images sample variability
Overview
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793392/image03.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793400/image11.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793402/image13.png)
Differing Saliency
- some nucleus can be easily seen
- some are barely distinguishable
- some have homogenous structure
- some have nucleous in the center
Overview
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793401/image12.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793389/image01.png)
Varying shape and size
- some are elliptical
- some are oblong
- some are large
- some are very small
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793409/image01.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793402/image13.png)
Overview
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793415/image05.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793427/image15.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793396/image07.png)
Varying level of nuclei attachment
- some images are very dense
- some are sparse
- some have large groups of glued nuclei
Pipeline
Input image
Preprocessing
Morphological pooling
Watershed ensemble generation
Thresholding
Morphological ensemble generation
Technical Details
Watershed pooling
Segmented mask
Preprocessing
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793563/image04_hematoxin.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793497/image04.png)
Hematoxylin
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793564/image04_eosin.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793556/image04_hematoxin.png)
Eosin
Background
Technical Details
Color deconvolution
Pipeline
Preprocessing
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793563/image04_hematoxin.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793497/image04.png)
Hematoxylin
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793564/image04_eosin.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793556/image04_hematoxin.png)
Eosin
Background
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793560/image04_hematoxin.png)
Hematoxylin Channel Extraction
Technical Details
Color deconvolution
Pipeline
Preprocessing
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793496/image04Reinhard.png)
Reinhard's Color Normalization
References:
- Reinhard, E., Adhikhmin, M., Gooch, B., Shirley, P.: Color transfer between images. IEEE Computer Graphics and Applications 21(5) (2001) 34–41
- Wang, Y., nd L. Wu, S.C., Tsai, S., Sun, Y.: A color-based approach for automated segmentation in tumor tissue classification. In: Proc. Conf. of the IEEE Engineering in Medicine and Biology Society. (2007) 6577–6580
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793499/image08Target.png)
Target image
Technical Details
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793497/image04.png)
Pipeline
Preprocessing
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793573/image04.png)
Hematoxylin Channel Extraction
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793496/image04Reinhard.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793497/image04.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793499/image08Target.png)
Reinhard's Color Normalization
Target image
Technical Details
Pipeline
Thresholding
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793573/image04.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793768/truth_binary_mask.png)
Possible algorithms:
- K-means clustering
-
Fuzzy c-means clustering
-
Gaussian mixture models
2. With spatial information
1. No spatial information
- Spatial information Fuzzy c-means clustering
-
Fast spatial distance weighted Fuzzy c-means clustering
-
Dictionary model
References:
- Zeng, Huang, Kang and Sang. Image segmentation using spectral clustering of Gaussian mixture models. Neurocomputing 144:346–356, 2014.
- Guo, Liu, Wu, Hong and Zhang. A New Spatial Fuzzy C-Means for Spatial Clustering. WSEAS Transactions on Computers 14:369-381, 2015.
- Dahl and Larsen. Learning Dictionaries of Discriminative Image Patches. In: Proc. British Machine Vision Conference. p.77, 2011.
- Hamed Shamsi and Hadi Seyedarabi, Member, IACSITA Modified Fuzzy C-Means Clustering with Spatial Information for Image Segmentation,International Journal of Computer Theory and Engineering, Vol. 4, No. 5, 2012.
Technical Details
Pipeline
Thresholding
K-means clustering
Parameters:
- number of color clusters
- number of most most intensive clusters to be classified as nuclei
Models
- k-means(3,1)
- k-means(4,2)
- k-means(6,3)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798524/image04_intensity_cluster_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798526/image04_intensity_cluster_2.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798527/image04_intensity_cluster_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798528/image04_intensity_cluster_4.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798529/image04_intensity_cluster_5.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798530/image04_intensity_cluster_6.png)
input
color clusters
1
6
5
4
3
2
Technical Details
Pipeline
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793573/image04.png)
Thresholding
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798524/image04_intensity_cluster_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798526/image04_intensity_cluster_2.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798527/image04_intensity_cluster_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798528/image04_intensity_cluster_4.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798529/image04_intensity_cluster_5.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798530/image04_intensity_cluster_6.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798531/image04_top_3_intensity_clusters_summed.png)
input
color clusters
top 3
most intensive clusters sum
1
6
5
4
3
2
4+5+6
K-means clustering
Parameters:
- number of color clusters
- number of most most intensive clusters to be classified as nuclei
Models
- k-means(3,1)
- k-means(4,2)
- k-means(6,3)
Technical Details
Pipeline
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793573/image04.png)
Thresholding
Gaussian Mixture Model
Parameters:
- number of color clusters
- number of most most intensive clusters to be classified as nuclei
- which of the distribution parameters (means, covariances,weights) should be updated
- which covariance option (diagonal,full,tied,spherical) should be used
Models
- GMM(3,1,{means,covariances},diagonal)
- GMM(6,3,{covariances},diagonal)
- GMM(6,3,{covariances},tied)
References:
- Zeng, Huang, Kang and Sang. Image segmentation using spectral clustering of Gaussian mixture models. Neurocomputing 144:346–356, 2014.
Technical Details
Pipeline
Thresholding
Spatial distance weighted fuzzy c-means clustering
Parameters:
- number of color clusters
- number of most most intensive clusters to be classified as nuclei
- fuzziness parameter
- spatial information importance
- number of neighbours
Models
- SDWFCM(3,1,2,0.9,8)
- SDWFCM(3,1,2,0.9,24)
- SDWFCM(3,2,2,0.9,24)
cluster center
spatial distance weighted funcion
weighted distance
membership probabilty
objective function
References:
- Hamed Shamsi and Hadi Seyedarabi, Member, IACSITA Modified Fuzzy C-Means Clustering with Spatial Information for Image Segmentation,International Journal of Computer Theory and Engineering, Vol. 4, No. 5, 2012.
Technical Details
Pipeline
Thresholding
Dictionary model
References:
- Dahl and Larsen. Learning Dictionaries of Discriminative Image Patches. In: Proc. British Machine Vision Conference. p.77, 2011.
Technical Details
Learning:
- seeding dictionaries of texture and label patches
- iterative clustering of patches according to label similarity
Segmentation:
- assigning the idealized label of the most similar cluster centroid
Pipeline
Thresholding
Parameters:
- patch size
- label similarity threshold
- percentage of seeding and for training
- number of training iterations
- adjustment coefficient for centroids of texture clusters
Models
- DM(1,0.4,0.01,0.1,5,0.05,0.5)
- DM(1,0.5,0.01,0.1,20,0.05,0.0)
- DM(2,0.6,0.01,0.1,5,0.05,0.0)
- DM(3,0.6,0.01,0.1,5,0.15,0.5)
References:
- Dahl and Larsen. Learning Dictionaries of Discriminative Image Patches. In: Proc. British Machine Vision Conference. p.77, 2011.
Technical Details
Dictionary model
Pipeline
Thresholding
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793742/image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793573/image04.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793768/truth_binary_mask.png)
Reality is far from perfect:
- leftover artifacts
- hollow shapes
- glued nuclei groups
Technical Details
Pipeline
Thresholding
Operations:
- dilation and erosion
- reconstruction by erosion
- closing and opening
- fill binary holes
Morphological Transformations
Technical Details
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793742/image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793768/truth_binary_mask.png)
Pipeline
Thresholding
Operations:
- dilation and erosion
- reconstruction by erosion
- closing and opening
- fill binary holes
Morphological Transformations
Problems:
- Sample diversity
- Diversity within one image
- "one size fits all" structuring element does not exist
Technical Details
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793742/image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793768/truth_binary_mask.png)
Pipeline
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1793742/image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798801/image_04_target_image_06_n_3_p_1_morpho_realization_1.png)
Morphological ensemble generation
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798803/image_04_target_image_06_n_3_p_1_morpho_realization_2.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798804/image_04_target_image_06_n_3_p_1_morpho_realization_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798805/image_04_target_image_06_n_3_p_1_morpho_realization_4.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798807/image_04_target_image_06_n_4_p_2_morpho_realization_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798809/image_04_target_image_06_n_6_p_3_morpho_realization_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798835/ensamble_image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798845/ensamble_image_04_target_image_06_n_3_p_1_covar_2_params_0_init_params_5_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798838/ensamble_image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1_2.png)
k-means(3,1)
k-means(4,2)
SDWFCM(3,2,2,0.9,24)
1
2
r
Morphological ensemble
Technical Details
Pipeline
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798801/image_04_target_image_06_n_3_p_1_morpho_realization_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798964/image_04_target_image_06_n_6_p_3_morpho_realization_4.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798804/image_04_target_image_06_n_3_p_1_morpho_realization_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798805/image_04_target_image_06_n_3_p_1_morpho_realization_4.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798807/image_04_target_image_06_n_4_p_2_morpho_realization_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798809/image_04_target_image_06_n_6_p_3_morpho_realization_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798835/ensamble_image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798845/ensamble_image_04_target_image_06_n_3_p_1_covar_2_params_0_init_params_5_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798838/ensamble_image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1_2.png)
k-means(3,1)
k-means(4,2)
SDWFCM(3,2,2,0.9,24)
1
2
r
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798939/morpho_probability_map.png)
Morphological ensemble
Morphological pooling
probabilty map
Technical Details
Pipeline
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798801/image_04_target_image_06_n_3_p_1_morpho_realization_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798964/image_04_target_image_06_n_6_p_3_morpho_realization_4.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798804/image_04_target_image_06_n_3_p_1_morpho_realization_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798805/image_04_target_image_06_n_3_p_1_morpho_realization_4.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798807/image_04_target_image_06_n_4_p_2_morpho_realization_1.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798809/image_04_target_image_06_n_6_p_3_morpho_realization_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798835/ensamble_image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798845/ensamble_image_04_target_image_06_n_3_p_1_covar_2_params_0_init_params_5_3.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798838/ensamble_image_04_target_image_06_n_6_p_3_covar_2_params_2_init_params_1_2.png)
k-means(3,1)
k-means(4,2)
SDWFCM(3,2,2,0.9,24)
1
2
r
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798939/morpho_probability_map.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798946/morpho_pooling_result.png)
Morphological ensemble
k-means clustering
Morphological pooling
probabilty map
Technical Details
Pipeline
nuclei detachment
Watershed ensemble generation
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798971/morpho_pooling_result.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798979/image04_mask_truth.png)
Watershed algorithm
- local maxima search space
- minimal distance between markers
- other
Technical Details
Pipeline
nuclei detachment
Watershed ensemble generation
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798971/morpho_pooling_result.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798979/image04_mask_truth.png)
Watershed algorithm
- local maxima search space
- minimal distance between markers
- other
Problems:
- Sample diversity
- Diversity within one image
- "one size fits all" parameter set does not exist
Technical Details
Pipeline
Watershed ensemble generation
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798971/morpho_pooling_result.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799581/image_04_loc_search_5_min_dist_2.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799587/image_04_loc_search_15_min_dist_10.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799591/image_04_loc_search_55_min_dist_20.png)
watershed realizations
thresholding output
Technical Details
Pipeline
Watershed ensemble generation
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798971/morpho_pooling_result.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799581/image_04_loc_search_5_min_dist_2.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799587/image_04_loc_search_15_min_dist_10.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799591/image_04_loc_search_55_min_dist_20.png)
watershed realizations
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799657/image_04_loc_search_5_min_dist_2.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799659/image_04_loc_search_15_min_dist_10.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799660/image_04_loc_search_55_min_dist_20.png)
thresholding output
edge realizations
Technical Details
Pipeline
Watershed pooling
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798971/morpho_pooling_result.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799657/image_04_loc_search_5_min_dist_2.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799659/image_04_loc_search_15_min_dist_10.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799660/image_04_loc_search_55_min_dist_20.png)
probability map
thresholding output
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799817/edges_prob_map.png)
edge realizations
Technical Details
Pipeline
Watershed pooling
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798971/morpho_pooling_result.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799657/image_04_loc_search_5_min_dist_2.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799659/image_04_loc_search_15_min_dist_10.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799660/image_04_loc_search_55_min_dist_20.png)
probability map
thresholding output
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799817/edges_prob_map.png)
modified thresholding output
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1800246/overimposed.png)
overimpose
edge realizations
Technical Details
Pipeline
Watershed pooling
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1798971/morpho_pooling_result.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799657/image_04_loc_search_5_min_dist_2.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799659/image_04_loc_search_15_min_dist_10.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799660/image_04_loc_search_55_min_dist_20.png)
probability map
thresholding output
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1799817/edges_prob_map.png)
modified thresholding output
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1800246/overimposed.png)
final segmentation mask
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1800395/final_segmentation.png)
watershed
overimpose
edge realizations
Technical Details
Pipeline
Our Team
Grzegorz Żurek
R&D Stermedia
Wroclaw University of Technology
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1797208/tfasz.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1782201/czakon_jakub.jpg)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1782199/piotr_giuedziun.jpg)
Jakub Czakon
R&D Stermedia
Piotr Giedziun
R&D Stermedia
Ph.D Witold Dyrka
R&D Stermedia
Wroclaw University of Technology
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1804241/piotr_krajewski.jpg)
Piotr Krajewski
CIO Stermedia
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1809590/michalblach.png)
Michał Błach
R&D Stermedia
Wroclaw University of Technology
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1794616/poziom-en.gif)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1804324/stermedia.png)
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1812144/fulawka.jpg)
MD Łukasz Fuławka
Patomorphology resident
Lower Silesian Oncology Center
![](https://s3.amazonaws.com/media-p.slid.es/uploads/382197/images/1804324/stermedia.png)
Contact us
info@stermedia.pl
jakub.czakon@stermedia.pl
grzegorz.zurek@stermedia.pl
We are looking forward to collaborating with you
Thank you for attention
Segmentation Algorithm
By Stermedia Sp. z o.o.
Segmentation Algorithm
presentation for MICCAI 2015
- 2,588