Bone Cancer Identification and Classification Using Hybrid Fuzzy Clustering With Deep Learning Classification

D. Joshua Jeyasekar, Dr.G. Arulselvi, V. Vedanarayanan and D. Poornima

In the recent past, bone cancer segmentation on bone CT image has been adopted as efficient diagnosis for clinical pathology. Efficient segmentation of bone cancer is the key procedure in radiotherapy planning. The CT imaging is faster and cost effective than MRI with low radiation. This paper presents novel bone cancer segmentation technique using hybrid clustering. To improve the quality of the CT image, the preprocessing steps are implemented. The Gabor wavelet filter is used to remove noise and scattered pixels are corrected by applying hexagonal sampled grid. The hybrid clustering segmentation incorporates K means and Fuzzy C Means (FCM) clustering. The automatic seed point is selected by K means clustering and high intensity gradient is estimated by applying FCM method. After segmenting the Region of Interest (ROI), the features are calculated using Gray Level Co-Occurrence Matrix (GLCM) and band let transform. The extracted features are given to deep learning to classify different stages of cancer. The experimental results are proven that the proposed algorithm is high efficient and accurate.

Volume 11 | 10-Special Issue

Pages: 88-98

DOI: 10.5373/JARDCS/V11SP10/20192779