Human Cancer Cell Line Classification
Experimental Web Server Version 0.1

To classify a cancer cell line image

(You can download the source MATLAB codes from here.)

Summary Cancer cell lines are widely used for research purposes in laboratories all over the world. Computer-assisted classification of cancer cells can alleviate the burden of manual labeling and help cancer research. In this paper, we present a novel computerized  method for cancer cell line image classification. The aim is to automatically classify 14 different classes of cell lines including 7 classes of breast and 7 classes of liver cancer cells. Directionally selective DT-CWT feature parameters are used because of their ability to characterize edges at multiple orientations which is the characteristic feature of carcinoma cell line images. A Support Vector Machine (SVM) classifier with radial basis function (RBF) kernel is employed for final classification. The proposed system can be used as a reliable decision maker for laboratory studies. This website gives interested scientists the opportunity to test our algorithm on their data.
Input Our software takes a JPG color image of the following 14 classes as input:

Breast cancer cell lines:
- BT-20
- Cama-1
- MDA-MB-157
- MDA-MB-361
- MDA-MB-453
- MDA-MB-468
- T47D

Liver cancer cell lines:
- Focus
- Hep40
- HepG2
- Huh7
- mv
- SkHep1

Image sizes should be equal to 3096-by-4140 or larger. The magnification factor of the image can be
chosen as 10x, 20x or 40x, respectively. The cancer type (breast or liver) can be given as additional information, if known. 
Output Our software displays the estimated class of the uploaded image and the confluency level.
Sample Example Images and datasets
Reference "Image Classification of Human Carcinoma Cells Using Complex Wavelet-Based Covariance Descriptors" by Furkan Keskin, Alexander Suhre, Kivanc Kose, Tulin Ersahin, Rengul Cetin-Atalay and A. Enis Cetin.
Contact Enis Cetin: cetin -at-    Rengul Cetin-Atalay: rengul -at-

Back to

This software was developed as part of the FP7 project MIRACLE.