Skip to content

Nuclei Segmentation and Feature Extraction Framework - C++ (ITK)

License

Notifications You must be signed in to change notification settings

humayun/NucleiSeg_FeatureExt_C

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

#Nuclei Segmentation and Feature Extraction Framework (Nuclei Morphology Analysis) Framework

This framework have different function to segment nuclei and compute nuclei features including (morphology, Intensity, Co-occurrence and Run-Length texture features). This framework is used to compute nuclei features of histopathology images of Breast lesions for PlosOne 2014 paper (Discriminate UDH and DCIS).

PlosOne Paper (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0114885)

This framework have different functions for computing nuclei segmentation and Feature extraction like:

1- Nuclei Segmentation only

2- Nuclei Segmentation and Feature Extraction

3- Feature Extraction only ( The segmented nuclei images should exist in the processing folder)

4- Nuclei Segmentation on Grayscale (Flouresence Images like Maximum Intensity Projection images of 3D Lightsheet Microsocpy)

5- Nuclei Feature extraction on ROI images of TCGA WSI

6- Nuclei Segmentation and Feature Extraction of ICPR MITOS Dataset 2012

7- Nuclei Feature Extraction of ICPR MITOS Dataset 2012

To build NucleiMorphology Program, two libraries are required: (ITK and Boost).

To run the NucleiMorphology Program, you need to pass following arguments:

-i “image folder location” (Input Directory: required)

-f numeric value ranging from 0 to 7 (Execution Mode: required)

• 0 – nuclei segmentation only,

• 1 – nuclei segmentation and feature extraction,

• 2 – nuclei feature extraction only,

• 3 – nuclei feature extraction for grayscale image (LSM),

• 5 – Mitosis segmentation and feature extraction,

• 6 – Mitosis feature Extraction only,

• 7 – feature extraction using selected color channels (4)

-l 128 (Grayscale Levels: optional, Default is 256; other values can be 128, 64, 32 or 16)

-c (NoOfColumns: optional, Default is 1, other values can be 2, 4, or 8, number of columns for computing texture features (Co-occurrence, Run-Length))

-r (NoOfRows: optioanl, Default is 1, other values can be 2,4, or 8, number of rows for computing texture features (Co-occurrence, Run-Length)

-p (Feature Computation Mode: optional, Default is region features; other value is patch based features.)

About

Nuclei Segmentation and Feature Extraction Framework - C++ (ITK)

Resources

License

Stars

Watchers

Forks

Packages

No packages published