TY  - JOUR
T1  - Digital Mammogram Segmentation and Tumour Detection Using Artificial Neural Networks
AU - , Y. Ireaneus Anna Rejani AU - , S. Thamarai Selvi 
JO  - International Journal of Soft Computing
VL  - 3
IS  - 2
SP  - 112
EP  - 119
PY  - 2008
DA  - 2001/08/19
SN  - 1816-9503
DO  - ijscomp.2008.112.119
UR  - https://makhillpublications.co/view-article.php?doi=ijscomp.2008.112.119
KW  - Discrete wavelet transform
KW  -fractal dimension analysis
KW  -region splitting
KW  -region filling
KW  -back propagation neural network and tumor detection
AB  - This study presents an algorithm which aims to assist the radiologist  towards fast detection and early diagnosis of breast cancer. It combines several image processing techniques like region splitting and region filling with the  Discrete Wavelet Transform  (DWT), artificial  intelligence techniques and artificial neural networks for detection of masses in mammograms. The AI techniques include fractal dimension analysis, dogs and rabbits algorithm and others. The fractal dimension analysis is used to find the roughness value, which is used to locate the region suspicious for cancer in the mammogram. The dogs-and-rabbits algorithm initiates the clustering. Region splitting and filling are used to segment the suspicious region. The Back Propagation neural network is applied at the end to determine whether a given mammogram is suspicious for cancer. The algorithm is verified by using mammograms from Mammographic Image Analysis Society database.
ER  -