ROUGH DRAFT authorea.com/115724
Main Data History
Export
Show Index Toggle 0 comments
  •  Quick Edit
  • Feature Extraction of a potential Melanoma

    Asymmetry

    Compactness

    Compactness of 2d region composed of square pixels(Bribiesca 1997)

    More compactness info(Jaworek-Korjakowska 2016)

    Compactness by perimeter ratios(Hernández 2016)

    Center of Mass

    Measure Array of radiuses from center of mass(Cudek):

    Calcualte $$SFA\alpha$$ ( Score for Axis ) for each potiential axis ( 0° - 179° )

    A: Is a main axis found above threshold?

    B: if yes, is the perpendicular axis also above threshold?

    if A & B -> Asymmetry Score = 0

    if A & !B -> Asymmetry Score = 1

    if !A -> Asymmetry Score = 2

    Comparisons of various methods

    Asymmetry Analysis "In ABCD analysis, out of other three parameters, asymmetry analysis predicts better and also it assists the clinicians to diagnose the melanoma before its proliferating stage."(Premaladha 2014)

    Border irregularity

    Fourier feature(Amelard 2013)

    Fractal Geometry

    Area and perimeter

    Irregularity index

    Borderling Function(Jaworek-Korjakowska 2015)

    Rotate the lesion image onto major axis

    Calculate Borderline Function

    Smoothing

    Calculate Border Irregularities from change in sign of slope of borderline function

    Implementation

    using Borderline Function

    https://github.com/alexgustafson/BATests/blob/master/Calculate%20Border%20Irregularity.ipynb

    Color

    Differential Structure(Alcon 2009)

    References

    1. E. Bribiesca. Measuring 2-D shape compactness using the contact perimeter. Computers & Mathematics with Applications 33, 1–9 (1997). Link

    2. Joanna Jaworek-Korjakowska, PawełK. Automatic Classification of Specific Melanocytic Lesions Using Artificial Intelligence. BioMed Research International 2016, 1–17 (2016). Link

    3. R. Santiago-Montero David Asael Gutiérrez Hernández. Border and Asymmetry measuring of skin lesion for diagnostic of melanoma using a perimeter ratio. Asian Journal of Computer Science and Information Technology 6 (2016).

    4. Pawel Cudek, Jerzy W. Grzymala-Busse, Zdzislaw S. Hippe.

    5. J. Premaladha, K.S. Ravichandr. Asymmetry Analysis of Malignant Melanoma Using Image Processing: A Survey. J. of Artificial Intelligence 7, 45–53 (2014). Link

    6. Robert Amelard. High-Level Intuitive Features (HLIFs) for Melanoma Detection. 85 (2013).

    7. Joanna Jaworek-Korjakowska. Novel Method for Border Irregularity Assessment in Dermoscopic Color Images. Computational and Mathematical Methods in Medicine 2015, 1–11 (2015). Link