An Imaging Method for Automated SkinLesion Segmentation using Statistical Analysis and Bit Plane Slicing

An Imaging Method for Automated SkinLesion Segmentation using Statistical Analysis and Bit Plane Slicing

Paper in proceedings (conference paper)

Melanoma can prove fatal if not diagnosed at early stage. Computer aided identification of diseases can equip normal people in performing a screening test of diseases. The accurate lesion segmentation plays a crucial role in correct diagnosis of skin diseases. This work proposes a skin lesion segmentation method using statistical analysis and bit plane slicing. Hole filling operation is competent enough to sufficiently reject the noisy pixels from the finally segmented image. The results of skin lesion segmentation obtained from the proposed algorithm has been compared with the annotated images available with the database. The results arepresented in form of overlapping score and correlation coefficient. An average overlapping score and correlation coefficient of 91.59% and 92.07%, respectively, is obtained from the proposed algorithm. Also, an image based performance analysis of segmented lesion has been done and an average sensitivity of 94.33% has been achieved. The results are convincing and suggests that the proposed work can be used for some real-time application.

Melanoma can prove fatal if not diagnosed at early stage. Computer aided identification of diseases can equip normal people in performing a screening test of diseases. The accurate lesion segmentation plays a crucial role in correct diagnosis of skin diseases. This work proposes a skin lesion segmentation method using statistical analysis and bit plane slicing. Hole filling operation is competent enough to sufficiently reject the noisy pixels from the finally segmented image. The results of skin lesion segmentation obtained from the proposed algorithm has been compared with the annotated images available with the database. The results arepresented in form of overlapping score and correlation coefficient. An average overlapping score and correlation coefficient of 91.59% and 92.07%, respectively, is obtained from the proposed algorithm. Also, an image based performance analysis of segmented lesion has been done and an average sensitivity of 94.33% has been achieved. The results are convincing and suggests that the proposed work can be used for some real-time application.

Medical Images; Dermoscopic Images; Skin Lesion;Melanoma; Statistical Analysis;

Medical Images; Dermoscopic Images; Skin Lesion;Melanoma; Statistical Analysis;

GUPTA, A.; ISSAC, A.; DUTTA, M.; ŘÍHA, K.

2019

09.02.2018

IEEE

GHAZIABAD, India

978-1-5386-0886-9

International Conference on "Computational Intelligence and Communication Technology"

1

5

5