Archives

  • 2022-05
  • 2022-04
  • 2021-03
  • 2020-08
  • 2020-07
  • 2020-03
  • 2019-11
  • 2019-10
  • 2019-09
  • 2019-08
  • 2019-07
  • br Fig Computer aided diagnosis CAD

    2022-05-07


    Fig. 2. Computer aided diagnosis (CAD) flow.
    Please cite this article as: H. A. Nugroho, Zulfanahri, E. L. Frannita et al., Computer aided diagnosis for thyroid cancer system based on internal and external characteristics, Journal of King Saud University – Computer and Information Scienceshttps://doi.org/10.1016/j.jksuci.2019.01.007
    4 H.A. Nugroho et al. / Journal of King Saud University – Computer and Information Sciences xxx (xxxx) xxx
    Cc
    N
    N
    ð Þ
    N yc;xc N
    yc;xc
    P
    c) Solidity is a concavity measurement of an object as calcu-
    S ¼ ObjectArea
    ConvexArea
    d) Compactness shows irregularity index of an object can be cal-culated in (4).
    Fig. 4. Speckle reduction bilateral filtering illustration (Paris et al., 2009).
    (mask-m) inside the image and let the curve evolved by itself by t-iteration to find minimum energy total described by the object. It was said to be minimum when the contour of an object inside the image was smooth. An example of performing active contour on an ultrasound image is illustrated in Fig. 5. The segmentation process was continued with morphological operation to increase the quality of the segmented image from active contour. Some obtained results still contained holes inside the object and the edge of the object was not as smooth as expected. Morphological operation worked with two-pixel arrays operation; the first array was the input image and the second array was the structuring element. To solve the aforementioned prob-lem, a structuring Veratridine was used based on filling and dilation operations (Kadir and Susanto, 2013). 
    Perimeter2
    e) Rectangularity indicates the similarity of an object to a rect-angular as formulated in (5).
    R ¼ ObjectArea ð5Þ
    RectangularArea
    f) Tortuosity is a measure of curve tortuous of an object as defined in (6).
    T ¼ 2 MajorAxisLength 6
    ObjectPerimeter
    2.3. Feature extraction
    To describe the important features on the thyroid ultrasound images, geometric and texture features were analysed since both features were related to the characteristics. There are nine geomet-ric features associated with shape, margin and orientation charac-teristics, whilst five texture features are associated with content and echogenicity characteristic.
    a) Convexity is a roughness measurement of an edge object as formulated in (1).
    Cx ¼ ConvexPerimeter ð1Þ
    ObjectPerimeter
    b) Circularity indicates the similarity of an object to a circle as defined in (2). 
    Fig. 6. GLCM process: (a) original image, (b) pixel combination, (c) number of pixel combination.
    Fig. 5. The illustration of active contour segmentation.
    Please cite this article as: H. A. Nugroho, Zulfanahri, E. L. Frannita et al., Computer aided diagnosis for thyroid cancer system based on internal and external characteristics, Journal of King Saud University – Computer and Information Scienceshttps://doi.org/10.1016/j.jksuci.2019.01.007
    H.A. Nugroho et al. / Journal of King Saud University – Computer and Information Sciences xxx (xxxx) xxx 5
    g) Ratio height-width is a comparison between the height and width of an object as described in (7).
    Rhw ¼ ObjectArea ð7Þ
    h) Dispersion is a measure of irregularity of an object based on length of the main chord ratio with the area of the object as defined in (8).
    D ¼ Majoraxislength ð8Þ
    ObjectArea
    i) Ratio aspect measures a comparison between the width of the object and the length of the object. The width here is dif-ferent from the width in point g. Point gis the shortest dis-tance between two points on the perimeter. The length is the line which perpendicular to the width. Ratio aspect is calculated using (9). 
    RatioAspect ¼ ObjectWidth ð9Þ
    ObjectLength
    Histogram is worked by calculating the spread of intensity value. Histogram has some features used to describe texture pat-tern of image. This study used eight features for representing tex-ture pattern. These were mean, standard deviation, skewness, energy, entropy, smoothness, variance and kurtosis (Jain, 1989). GLCM, which stands for grey level co-occurrence matrix is used to represent texture patterns in first order using statistical opera-tion based on pixel values. GLCM expresses the correlation between two pixels from various angular directions (Haralick et al., 1973). GLCM is initiated by calculating the number of related pixels. An example of GLCM process is illustrated in Fig. 6. This pro-cess produces a framework matrix. The framework matrix is used to produce GLCM matrix by counting it up with the transpose value.
    Table 1
    Fig. 7. Laws’ vector multiplication of E5 and L5 to produce energy maps.
    Table 2
    Nine Laws’ energy maps.
    R5R5 Fig. 9. Multilayer perceptron (MLP) architecture.