Novel ABCD Formula to Diagnose and Feature Ranking of Melanoma

A prototype of skin cancer detection system for melanoma diagnoses in early stages is very important. In this paper, a novel technique is proposed for Skin malignant growth identification based on feature parameters, color shading histogram, to improve the diagnosis method by optimizing the ABCD formula. Features are extracted like Shape, Statistical, GLCM texture, Color, Wavelet transform, Texture. Once the features are extracted we found the most prominent features by assigning a rank. We have calculated parameters such as sensitivity, specificity, accuracy for checking the imperceptibility and robustness of the proposed approach. Also, Correlation analysis is made between traditional and proposed TDS equation using Karl Pearson’s method.

@article{M2019,
  title     = {Novel ABCD Formula to Diagnose and Feature Ranking of Melanoma},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {10},
  number    = {1},
  year      = {2019},
  publisher = {The Science and Information Organization},
  author    = {Reshma M and B. Priestly Shan},
  doi       = {10.14569/IJACSA.2019.0100111},
  url       = {https://doi.org/10.14569/IJACSA.2019.0100111}
}