{"id":56017,"date":"2024-03-20T10:34:53","date_gmt":"2024-03-20T10:34:53","guid":{"rendered":"https:\/\/biomedpharmajournal.org\/?p=56017"},"modified":"2024-04-02T04:20:33","modified_gmt":"2024-04-02T04:20:33","slug":"human-epithelial-cell-image-analysis-and-segmentation-using-threshold-based-fusion-technique","status":"publish","type":"post","link":"https:\/\/biomedpharmajournal.org\/staging\/vol17no1\/human-epithelial-cell-image-analysis-and-segmentation-using-threshold-based-fusion-technique\/","title":{"rendered":"Human Epithelial Cell Image Analysis and Segmentation using Threshold Based Fusion Technique"},"content":{"rendered":"\n

Introduction<\/strong><\/p>\n\n\n\n

The\nrapid development of new technologies in the field of image processing has\nhelped in discovering human abnormalities at an early stage and is faster than\nmanual methods. Image analysis and segmentation have an essential role in the\nmedical field and help in the early detection of diseases in critical health conditions.\nDue to the structural complexity and type of the cell image, radiologists may\nface difficulties in extracting the necessary features of the image, 1<\/sup>\nwhich are crucial in treating the patient. Therefore, correct and meaningful\nsegmentation is a challenging task and is required for further process. <\/p>\n\n\n\n

The\nmost used substrate for finding the antinuclear antibody (ANA) is the human\nepithelial type2 (HEp-2) cell samples, which are a natural protein array with\nplenty of antigens. The discovery of ANA in human serum is a vital transmission\ninstrument for connective tissue syndromes, and immune \ufb02uorescent (IIF) is the\nreference method for ANA testing. Indirect immune \ufb02uorescent (IIF-ANA) test is\nbecoming increasingly crucial for analyzing the cell image because of its ease\nand inexpensiveness 1, 2<\/sup>. The attributes of epithelial type-2\n(HEp-2) cells are an oval shape with merged properties of connected tissues,\nwhich help in microscopic observation and provide better result in detecting\nthe number of antibodies with better resolution. Therefore, cell image\nsegmentation is an essential task that examines the appropriate biological\nfeatures such as cell type, shape, group, etc. The experimental studies are\nfocused mainly on (HEp-2) cell image and aim to identify and segment the cells\nbased on ground truth data.<\/p>\n\n\n\n

Several\ncomputer vision methods have been developed for a variety of applications in\nimage processing. The first phase of computer vision includes pre-processing\nand image segmentation.3<\/sup> In this stage dissimilar objects are\nidentified and separated from the background. The next part is feature\nextraction, where object are measured. The computing of image features to\nquantitatively evaluate some significant features of object. These clusters are\ncombined to create a feature vector during feature mining. 4, 5<\/sup> The\nnext level is classification. At this point, the output is only a decision to\nconclude the class of each object. The structural study is an explanation of\nimages to properly recognize and judge the vital details of images. 6 – 15<\/sup><\/p>\n\n\n\n

Thresholding<\/strong><\/p>\n\n\n\n

Thresholding\nis the fundamental way to do image segmentation. The grayscale image is taken\nas input data, and the output image will be in the form of binary. The simplest\nthresholding 3<\/sup> method substitutes every pixel in an image with a\ndark pixel if the image intensity {\\display style I {i, j}} Ii, j is less\nthan definite static constant T (that is, Ii, j < T {\\display\nstyle I {i, j} < T}), or a bright pixel if the image intensity is more than\nthat constant. <\/p>\n\n\n\n

Global threshold<\/strong><\/p>\n\n\n\n

The histogram is a\ngraphical way of representing the image intensity. It reveals two points for an\nindication from the background and the foreground, respectively. Manual\nthresholding 3<\/sup> involves intensity value (threshold) such that total\npixels having an intensity value lesser than the threshold fit to one segment,\nand the rest of the part fits the other. Global thresholding provides better\nresults only when the degree of intensity separation between the two points in\nthe image. It is an unsophisticated segmentation selection 4<\/sup>.<\/p>\n\n\n\n\n\n
\"\"<\/td>\n\n

Figure 1: <\/strong>Graphical representation of histogram data.<\/strong><\/p>\n

<\/p>\n

Click here to view Figure<\/a><\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n

\"\"<\/figure>\n\n\n\n

Otsu\u2019s threshold<\/strong><\/p>\n\n\n\n

Image segmentation splits an image into its subareas and is a crucial task in mining image features, especially in medical image analysis and segmentation approaches. Due to its understanding of perception and ease, the thresholding technique is important and essential part of diagnosing medical images. 5, 6<\/sup> Usually, the histogram-dependent threshold choice technique, and \u201cthe maximum between-class variance method proposed by the Japanese scholar Otsu used\u201d. The selection of principle threshold criteria technique reflects the largest interclass difference between the object and background. Similarly, by varying the Otsu\u2019s threshold level is either 0 or 1 to get an FCM threshold. Thus, in the analysis of general as well as medical images, this technique provides better results compared to some of the popular methods. 11, 12<\/sup> The algorithm steps are given by <\/p>\n\n\n\n

Initial estimation of T<\/p>\n\n\n\n

Segmentation using of T<\/p>\n\n\n\n

g1 pixel brighter than T<\/p>\n\n\n\n

g2 pixel darker than T<\/p>\n\n\n\n

[computation of average Intensities m1 and m2 of g1 and g2]\n\n\n\n

\"\"<\/figure>\n\n\n\n

if [T-Tnew]>\u2206Tbacktostep 2, otherwise stop.<\/p>\n\n\n\n

Adaptive threshold<\/strong><\/p>\n\n\n\n

In an adaptive threshold, the threshold value is calculated for a particular region, thus it varies from the threshold rate at every pixel position based on the adjacent pixel values. 6<\/sup> To estimate the threshold T (x, y) i.e. the threshold value at pixel position (x, y) in the image.13-16<\/sup><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Multilevel threshold<\/strong><\/p>\n\n\n\n

Multilevel thresholding is the most applicable and popular image segmentation technique. 7<\/sup> Here, thresholding is assigned to segment a gray-level image into some discrete areas. Here, more than one threshold criterion is chosen to analyze the given image. 10<\/sup><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Optimal threshold<\/strong><\/p>\n\n\n\n

Choosing a threshold in the valley between two overlapping peaks classifies the pixels inappropriate way. The selection of optimal threshold and the optimal regulation can be done inevitably by a similar learning process and is explicitly for the optimal selection of threshold used in predications. The directions from examples and counterexamples have been projected by reducing an entropy function. 8<\/sup> The rules have the procedure of disjunctions of conjunctions of predicates. Thus, such a method can study commands from cases, and depends on an optimality principle. 9, 10<\/sup><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

pb (z), p0 (z), prob. distributions of background, object pixels. <\/p>\n\n\n\n

\u03bcb, \u03bc0 : the means of the distributions. <\/p>\n\n\n\n

\u03c3b, \u03c30 : the standard deviations of the distributions. <\/p>\n\n\n\n

Pb, Po : the a-priori probabilities of background, object pixel. 17<\/sup><\/p>\n\n\n\n\n\n
\"\"<\/td>\n\n

Figure 2: Optimal threshold selection in between the overlapping peaks.<\/strong><\/p>\n

<\/p>\n

Click here to view Figure<\/a><\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n

K-means clustering<\/strong><\/p>\n\n\n\n

This algorithm is a simple and effective unsupervised technique, which can be used for various applications like abnormality detection, pattern analysis, image scene understanding, etc. It classifies a sample set Y (y1, y2, y3\u2026. yn<\/sub>) into groups with the endeavor at minimizing an objective function J is given by, 18<\/sup><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

where, n represents the quantity of samples and ci<\/sub> is the cluster center, where, \u2016yj<\/sub>(i) <\/sup>-ci<\/sub> \u20162<\/sup> is the \u201cEuclidean distance measure between a data point yj<\/sub> and the cluster center ci<\/sub>\u201d. 19<\/sup><\/p>\n\n\n\n

Algorithm steps<\/strong><\/p>\n\n\n\n

Compare the distributed pixel intensity of the objects. <\/p>\n\n\n\n

Initialize the centroids with K random intensities.<\/p>\n\n\n\n

Recurrence the step till each cluster get labeled.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Here, N is the sample number of the ith<\/sup> cluster.<\/p>\n\n\n\n

Cluster the points based on distance of their intensities from the centroid intensities.<\/p>\n\n\n\n

Compute the new centroid for each of the clusters. 20<\/sup><\/p>\n\n\n\n

Fuzzy c-means clustering<\/strong><\/p>\n\n\n\n

FCM\nalgorithm for detecting optimal partition of cluster points is one of the\npopular techniques for analysis of both cell and MRI images. It splits the data\npoints into subareas named cluster groups. Here the clustering points are\nautomatically assigned after the pre-processing stage, and the associate function\nis the curve that defines how each point in the input space is mapped to the\ndegree of associated value between 0 and 1. During the iteration process, at\neach cycle, optimized cluster points are generated. 18, 21, 22<\/sup><\/p>\n\n\n\n

Where, A represents the image and A = {Z (A, B), 0 < A \u2264 M, 0 < B \u2264 N}, and M*N denotes the cluster points.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

where, Y= {Y0, Y1, Y2,\u2026. Yc-1} signifies the data points<\/p>\n\n\n\n

X = [\u03bcc (p, q)] is the matrix, <\/p>\n\n\n\n

\u2018n\u2019 represents the constant.<\/p>\n\n\n\n

\u03bcc<\/sub> (p, q) is the associated data points of Z (A, B) at the Cth<\/sup> cluster,<\/p>\n\n\n\n

\u201cdc <\/sub>(p, q) is the Euclidian distance of the data point Z (A, B) from the Cth<\/sup> cluster\u201d. 21<\/sup><\/p>\n\n\n\n

Clusters number K (2 \u2264 K < M*N) with original clustering matrix values are labeled. <\/p>\n\n\n\n

The\ndata points Yc<\/sub>, and fuzzy matrix Xc<\/sub> (p, q) are fixed by\nutilizing equations [9], [10], correspondingly. <\/p>\n\n\n\n

The procedure is recurring till the quantities varying among two iterations and is not more than \u03b5. 22, 23<\/sup><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n\n\n\n

The rest of this paper is structured as, proposed work in Section 2, the flowchart, and algorithm of the proposed method are presented in section 3, the result are given in section 4, and the conclusions in section 5.<\/p>\n\n\n\n

Proposed method<\/strong><\/p>\n\n\n\n

Firstly, the input\ncell image is transformed to grey scale, and a Gaussian filter is employed to\neliminate noise contamination in the cell image. 8<\/sup> A denoised image\nis then thresholded by an automated higher threshold value in conjunction with\nbinary smoother. However, the thresholded pixel values are fewer than the\nimage’s mean frequency. The resulting image is processed with a “remove\nborder” operation, and the thresholded image is equalized using an\nadaptive histogram. The picture is simultaneously denoised and segmented using\nan FCM (fuzzy C-means) based clustering approach. Finally, combining these two\nwith the fusion procedure results in a new segmented image. The resultant image\nis more suitable to ground truth data, and is compared with existing\nthresholding algorithms. 12<\/sup><\/p>\n\n\n\n

Gaussian filter<\/strong><\/p>\n\n\n\n

The Gaussian filter is a type of linear smoothing filter in which the weights are chosen according to the Gaussian function procedure for removing noise from a normal distribution. Two-dimensional discrete Gaussian zero mean-function is employed as a smoothing filter in image processing. <\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Higher Threshold<\/strong><\/p>\n\n\n\n

Bernsen’s locally adaptive binarization approach is utilized to extract the adhesive cell image feature from a grayscale picture. It is also tested for various neighborhood values and contrast limitations.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

(Zlow<\/sub>) is the lowest and (Zhigh<\/sub>) is the highest gray level pixel value in a square r x r neighborhood centered at pixel (x, y), <\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

If the contrast measure C (x, y) < l, the neighborhood is made up of only one class, foreground or background. Furthermore, r and l values altered based on the image regions employed.<\/p>\n\n\n\n

Binary smoother<\/strong><\/p>\n\n\n\n

Binary smoothing is a method for smoothing\ndata or filtering noisy data. The data is first categorized, and the sorted\nvalues are distributed across many segments or cells. The technique conducts\nlocal smoothing since they refer to a neighborhood of data. Smoothing by binary\nmeans replaces each value in a binary with the binary mean value.<\/p>\n\n\n\n

Adaptive histogram<\/strong><\/p>\n\n\n\n

The image is splits into discrete blocks, and a histogram is computed for each segment. Thus, adaptive histogram equalization computes many histograms, each of which corresponds to a different part of the picture. Consequently, it improves the local contrast and edge definitions in all segregated sections of the image.<\/p>\n\n\n\n

Material and methods <\/strong><\/p>\n\n\n\n

Flow chart<\/strong><\/p>\n\n\n\n\n\n
\"\"<\/td>\n\n

Figure 3: <\/strong>Schematic flowchart of the proposed method<\/strong>.<\/strong><\/p>\n

<\/p>\n

Click here to view Figure<\/a><\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n

Algorithm steps<\/strong><\/p>\n\n\n\n

Input: Cell image to\nbe segmented<\/p>\n\n\n\n

Output: Segmented image<\/p>\n\n\n\n

A cell image is converted to a gray-scale image.<\/p>\n\n\n\n

Image denoising is performed with the Gaussian filter to diminish the noise contamination in the cell image.<\/p>\n\n\n\n

Higher_threshold value is used to extract the required features from a denoised image, with binary smoother.<\/p>\n\n\n\n

Segmented image is processed by adaptive histogram, which is more accurate and noise-free.<\/p>\n\n\n\n

The combination of FCM clustered image along with the segmented image is fused to obtain the final target image.<\/p>\n\n\n\n

Results and Discussions<\/strong><\/p>\n\n\n\n

Performance matrices<\/strong><\/p>\n\n\n\n

The proposed algorithm is coded and executed in (MATLAB R2020a). The cell image (Hep-2) and the corresponding reference data is extractedfrom SNPHep-2 Dataset. The results are compared with existing methods and are found to be better.<\/p>\n\n\n\n

Accuracy <\/strong><\/p>\n\n\n\n

It specifies the perfect class of the image pixels. The measure of accuracy 24<\/sup> is given by<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Pixels which are perfectly incorporated into the given class are represented by TP and TN represents the pixels which are not belonging to the specified class. FN is incorrectly projected pixels, which are not fitting to the specified class.<\/p>\n\n\n\n

Sensitivity<\/strong><\/p>\n\n\n\n

Sensitivity is calculated through positive predicted and negative predicted values, which is the ratio of true outcome of all segmented results. 24<\/sup> <\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Dice co-efficient<\/strong><\/p>\n\n\n\n

Dice-coefficient examines the identical things in between the segmented data and the reference data, which is retrieved from manual segmentation. Dice-coefficient of segmented image 24, 25 <\/sup>is given by<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

where, |M|and |N|cardinalities of two sets.<\/p>\n\n\n\n

Jaccard-coefficient<\/strong><\/p>\n\n\n\n

Jaccard-coefficient is as same as the dice-coefficient, which calculates the commonness in-between ground truth data with different segmentation methods employed in this paper and 24, 25<\/sup> is given by<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Experimental results <\/strong><\/p>\n\n\n\n\n\n
\"\"<\/td>\nFigure 4: a. Input cell raw image, b. Ground truth image, c. Enhanced image, d. Grayscale image<\/strong>

<\/p>\n

Click here to view Figure<\/a><\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\"\"<\/td>\n\n

Figure 5: Segmented by: a. Global_ threshold, b. Otsu\u2019s_threshold, c. Adaptive_threshold, d. Multilevel_threshold.<\/strong><\/p>\n

<\/p>\n

Click here to view Figure<\/a><\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n
\"\"<\/td>\n\n

Figure 6: Segmented by: a. K-means_clustering, b. FCM_clustering, c. FCM_threshold, d. Higher_threshold, e. Proposed method.<\/strong><\/p>\n

<\/p>\n

Click here to view Figure<\/a><\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n

Table 1: Comparison of Hep-2 cell image segmentation result.<\/strong><\/p>\n\n\n\n\n\n\n\n\n\n
\n

Cell image<\/strong><\/p>\n<\/td>\n

\n

Global_T<\/strong><\/p>\n<\/td>\n

\n

Otsu\u2019s_T<\/strong><\/p>\n<\/td>\n

\n

Adaptive_T<\/strong><\/p>\n<\/td>\n

\n

Multi-<\/strong><\/p>\n

Level_T<\/strong><\/p>\n<\/td>\n

\n

Fcm-C<\/strong><\/p>\n<\/td>\n

\n

K-means<\/strong><\/p>\n<\/td>\n

\n

Fcm_T<\/strong><\/p>\n<\/td>\n

\n

Higher_T<\/strong><\/p>\n<\/td>\n

\n

Proposed method<\/strong><\/p>\n<\/td>\n<\/tr>\n

\n

Accuracy  in %<\/strong><\/p>\n<\/td>\n

\n

97.39<\/p>\n<\/td>\n

\n

95.67<\/p>\n<\/td>\n

\n

86.65<\/p>\n<\/td>\n

\n

95.67<\/p>\n<\/td>\n

\n

95.67<\/p>\n<\/td>\n

\n

94.89<\/p>\n<\/td>\n

\n

97.34<\/p>\n<\/td>\n

\n

97.41<\/p>\n<\/td>\n

\n

97.66<\/p>\n<\/td>\n<\/tr>\n

\n

Sensitivity in %<\/strong><\/p>\n<\/td>\n

\n

92.01<\/p>\n<\/td>\n

\n

85.82<\/p>\n<\/td>\n

\n

85.82<\/p>\n<\/td>\n

\n

85.82<\/p>\n<\/td>\n

\n

85.70<\/p>\n<\/td>\n

\n

91.14<\/p>\n<\/td>\n

\n

91.42<\/p>\n<\/td>\n

\n

90.60<\/p>\n<\/td>\n

\n

95.68<\/p>\n<\/td>\n<\/tr>\n

\n

Dice Coefficient in %<\/strong><\/p>\n<\/td>\n

\n

93.96<\/p>\n<\/td>\n

\n

89.20<\/p>\n<\/td>\n

\n

89.20<\/p>\n<\/td>\n

\n

89.20<\/p>\n<\/td>\n

\n

89.17<\/p>\n<\/td>\n

\n

88.07<\/p>\n<\/td>\n

\n

93.46<\/p>\n<\/td>\n

\n

93.98<\/p>\n<\/td>\n

\n

94.44<\/p>\n<\/td>\n<\/tr>\n

\n

Jaccard Coefficient in %<\/strong><\/p>\n<\/td>\n

\n

88.67<\/p>\n<\/td>\n

\n

80.50<\/p>\n<\/td>\n

\n

80.50<\/p>\n<\/td>\n

\n

80.50<\/p>\n<\/td>\n

\n

80.45<\/p>\n<\/td>\n

\n

78.68<\/p>\n<\/td>\n

\n

87.20<\/p>\n<\/td>\n

\n

87.93<\/p>\n<\/td>\n

\n

89.47<\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n\n\n\n
\"\"<\/td>\n\n

Figure 7: Performance evaluation of suggested study with existing techniques.<\/strong><\/p>\n

<\/p>\n

Click here to view Figure<\/a><\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n

Conclusion<\/strong><\/p>\n\n\n\n

A unique threshold-based image segmentation approach is suggested in this paper. The proposed model’s performance is compared to manual and current segmentation approaches, as well as the pre-processing procedure. However, instead of current segmentation strategies, the suggested method delivers superior segmentation by integrating the FCM-based clustering technique with image segmented by higher thresholding, which is one of the study’s important advantages. Furthermore, classic threshold-based segmentation approaches such as Otsu\u2019s, multi-level, and FCM thresholding methods, suffer, loss of intrinsic information after segmenting the cells, whereas global and adaptive thresholding give superior segmentation, but the threshold criteria must be set manually. Although unsupervised clustering techniques such as K-means and FCM are popular, the findings of both approaches do not meet the acceptable level throughout the assessment, particularly for adhesive cell segmentation. As a result, the proposed method has an edge and is useful in segmenting the adhesive Hep-2 cell image without sacrificing essential information at the fusion level. While examining the biological aspects of the Hep-2 cell image, radiologists will find the entire research quite useful. The simulation output of the proposed study is more effective and efficient in terms of computing. It is discovered that performance metric\u2019s sensitivity parameter improves by around 3%, and the segmented image’s effectiveness is assessed in terms of accuracy, sensitivity, dice-coefficient, and Jaccard-coefficient. The main contribution of this work is the improvement of performance measure, which is supported in all of these factors and is clear from the outcomes of the proposed algorithm.<\/p>\n\n\n\n

Acknowledgement<\/strong><\/p>\n\n\n\n

Standard test image samples used in this work were extracted from SNPHEp-2 Dataset, (https:\/\/staff.itee.uq.edu.au\/lovell\/snphep2\/).<\/p>\n\n\n\n

Conflict of interest<\/strong><\/p>\n\n\n\n

There is no conflict of interest.<\/p>\n\n\n\n

Funding Source<\/strong><\/p>\n\n\n\n

There are no funding sources<\/p>\n\n\n\n

References<\/strong><\/p>\n\n\n\n

  1. Rintu Maria Thomas, Jisha John, A review on cell detection and segmentation in microscopic images, 2017 international conference on circuits, power and computing technologies [iccpct], IEEE xplore: 19 october 2017, doi: 10.1109\/iccpct.2017.8074189.
    CrossRef<\/a><\/li>
  2. Simone Tonti Santa Di Cataldo Andrea Bottino <\/a>Elisa Ficarra, An automated approach to the segmentation of HEp-2 cells for the indirect immunofluorescence ANA test, Computerized Medical Imaging and Graphics, Volume 40, March 2015, Pages 62-69, https:\/\/doi.org\/10.1016\/j.compmedimag.2014.12.005.
     CrossRef <\/a><\/li>
  3. S.Sudha, G.R.Suresh and R.Sukanesh, Speckle noise reduction in ultrasound images by wavelet thresholding based on weighted variance. doi:10.7763\/ijcte.2009.v1.2.
     CrossRef <\/a><\/li>
  4. Huiping Yang, Haoyun Kuang, Xu Qin, and Guanyu Cai, Global threshold prediction of multicarrier multipactor with time distribution and material coefficients he wen, member, IEEE, doi: 10.1109\/temc.2017.2763955.
     CrossRef <\/a><\/li>
  5. J. Gholampour, A.a. Pouyan, The segmentation of FMI image layers based on FCM clustering and otsu\u2019s thresholding, international journal of scientific & engineering research volume 3, issue 3, march-2012, ISSN 2229-5518.<\/li>
  6. Panelshijian Chenpengjianshang, An adaptive method for threshold of recurrence quantification analysis based on sax, doi.org\/10.1016\/j.cnsns.2019.105061.<\/li>
  7. Amit Gurung, Sangyal Lama Tamang, Image segmentation using multi-threshold technique by histogram sampling, september 2019, project: digital image processing.<\/li>
  8. Panelb.Wangab, l.Chenaj.Chengc, New result on maximum entropy threshold image segmentation based on P system. doi.org\/10.1016\/j.ijleo.2018.02.062.<\/li>
  9. Chunyuan Wan; Mingquan Ye; Chuanwen Yao; Changrong Wu, Brain MR image segmentation based on gaussian filtering and improved FCM,  2017 10th international congress on image and signal processing, biomedical engineering and informatics (cisp-bmei), IEEE xplore: 27 february 2018, doi: 10.1109\/cisp-bmei.2017.8301978.
     CrossRef <\/a><\/li>
  10. Thresholding (Jain et al., Sections 3.2.1, 3.2.2, Petrou et al., Chapt 7).<\/li>
  11. Xiaoqiang Ji, Yang Li, Jiezhang Cheng, Yuanhua Yu*, Meijiao Wang, Cell image segmentation based on an improved watershed algorithm, 2015 8th international congress on image and signal processing (cisp 2015), 978-1-4673-9098-9\/15\/$31.00 \u00a92015 IEEE.<\/li>
  12. N.Valliammal, Dr.S.N.Geethalakshmi, Leaf image segmentation based on the combination of wavelet transform and k-means clustering. (ijarai) international journal of advanced research in artificial intelligence, vol. 1, no. 3, 201237 | page www.ijarai.thesai.org, Vol. 1, No. 3, 2012, doi:10.14569\/ijarai.2012.010307.
     CrossRef <\/a><\/li>
  13. M. Petrou and P. Bosdogianni, Image processing the fundamentals, wiley, uk, 2004.r. c. gonzalez and r.e. woods, digital image processing 2nd edition, prentice hall, new jersey, 2002.
     CrossRef <\/a><\/li>
  14. Mantosh Biswas* and Hari Omindian, A new soft-thresholding image denoising method, 2nd international conference on communication, computing & security [icccs-2012] . school of mines, dhanbad-826004, jharkand, india. Procedia Technology 6 (2012) 10 \u2013 15, doi: 10.1016\/j.protcy.2012.10.002.
    CrossRef <\/a><\/li>
  15. B. K. Shreyamsha Kumar, Image denoising based on non-local means filter and its method noise thresholding. doi 10.1007\/s11760-012-0389-y. <\/li>
  16. Grace Chang, bin yu, martin vetterli, Adaptive wavelet thresholding for image denoising and compression. February 2000, IEEE transactions on image processing 9(9):1532-46, doi:10.1109\/83.862633.
     CrossRef <\/a><\/li>
  17. H.D. Cheng, X.H. Jiang, Jingli, Color image segmentation based on homogram thresholding and region merging. Wang department of computer science, utah state university, logan, ut 84322-4205, usa received 31 january 2000; received in revised form 29 january 2001; accepted 29 january 2001. <\/li>
  18. Youguo Li, Haiyan Wu, A clustering method based on k-means algorithm, 2012 international conference on solid state devices and materials science, elsevier physics procedia 25 (2012) 1104 \u2013 1109. december 2012physics procedia 25:1104-1109, doi:10.1016\/j.phpro.2012.03.206.
     CrossRef <\/a><\/li>
  19. B. Padmapriya, M. S. Sangeetha, G. Ramya Priya Nandhini, T. T. Anusha Devi. Detection of malarial parasites using image processing techniques from blood smear slides, RJPT, volume – 11, issue – 10, year – 2018, doi: 10.5958\/0974-360X.2018.008053.
     CrossRef <\/a><\/li>
  20. Jianwei Liu1, a, Lei Guo1, b *, An improved k-means algorithm for brain MRI image segmentation, 3rd international conference on mechatronics, robotics and automation (ICMRA 2015). Available Online April 2015, https:\/\/doi.org\/10.2991\/icmra-15.2015.210.
     CrossRef <\/a><\/li>
  21. Abhishek Bala, *, Minakshi Banerjeea, Amlan Chakrabartib, Punit Sharmac, MRI brain tumor segmentation and analysis using rough-fuzzy c-means and shape-based properties, journal of king saud university, computer and information sciences, 1319-1578, https:\/\/doi.org\/10.1016\/j.jksuci.2018.11.001.
     CrossRef <\/a><\/li>
  22. Yogita K. Dubey and Milind M. Mushrif, FCM clustering algorithms for segmentation of brain MR images, hindawi publishing corporation advances in fuzzy systems, volume 2016, article id: 3406406, 14 pages, http:\/\/dx.doi.org\/10.1155\/2016\/3406406.
     CrossRef <\/a><\/li>
  23. Akila Victor, Muhammad Rukunuddin Ghalib, Detection of skin cancer cells \u2013a review, RJPT,  Volume – 10, Issue – 11, Year – 2017, DOI: 10.5958\/0974-360X.2017.00742.9. 
     CrossRef <\/a><\/li>
  24. Karlijn, J. van, Stralen1 vianda, S. Stel1 johannes, B. Reitsma2 friedo, W. Dekker13 carminezoccali4 kitty,  J.Jager1. Diagnostic methods i: sensitivity, specificity, and other measures of accuracy, under an Elsevier user license,https:\/\/doi.org\/10.1038\/ki.2009.92.
     CrossRef <\/a><\/li>
  25.  Jeroenbertels j.b. and T.E. Optimizing the dice score and jaccard index for medical image segmentation: theory & practice, have contributed equally to this work.11esat, center for processing speech and images, kuleuven, Belgium.<\/li><\/ol>\n","protected":false},"excerpt":{"rendered":"

    Introduction The rapid development of new technologies in the field […]<\/p>\n","protected":false},"author":15,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[113],"tags":[],"class_list":["post-56017","post","type-post","status-publish","format-standard","hentry","category-vol17no1"],"_links":{"self":[{"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/posts\/56017","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/users\/15"}],"replies":[{"embeddable":true,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/comments?post=56017"}],"version-history":[{"count":5,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/posts\/56017\/revisions"}],"predecessor-version":[{"id":57487,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/posts\/56017\/revisions\/57487"}],"wp:attachment":[{"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/media?parent=56017"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/categories?post=56017"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/tags?post=56017"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}