Global Journal of Science Frontier Research, A: Physics and Space Science, Volume 22 Issue 1

cluster to be able to well identify the tumor only resulted in the whole image becoming black. Figure 18: The processed image of sample 7 and its histogram From figure 18, it can be seen that the mass was seen when the fourth cluster was applied, the histogram also indicated that the tumor has gotten to the mass stage but smaller as compared to the one in figure 19. Figure 19: The processed image of sample 8 and its histogram From figure 19, the tumor was identified when the cluster was increased to 4 the when the histogram for the image indicate that the vibrant colors showed that the tumor was large as compared to the other tumors in the other images. The experimental results suggest that the introduced method for defect segmentation in this research is robust because it can accurately segment the cancerous part with the breast region, background and the blurred region of interest (ROI) boundary[19]. Finally, the detection accuracy was estimated and compared the performance with previous similar research works emphasizing the detection accuracy. The results obtained are also in support of anticipation with the findings and diagnosis by the radiologist of Mammocare Ghana of cancer research. The accuracy of detection has increased. IV. C onclusion In this work, the cancerous mammography segmentation of mammograms using K-means cluttering based on L*a*b* colour space was proposed and evaluated. The proposed approach used the K- means clustering technique for segmenting mammogram image four clusters. Mammograms images were used for the experimental observations and the introduced method was evaluated considering a cancerous mammogram as a case study. Experimental results suggest that the proposed approach is capable of accurately segmenting the tumor(mass) area of mammograms present in images. K-means based tumor segmentation approach is to also segment the cancerous area of the mammogram. R eferences R éférences R eferencias 1. Abdel-Mottaleb, M., Carman, C, S., Hill, C, R., & Vafai, S. (1996). “Locating the Boundarybetween the Breast Skin Edge and the Background in Digitized Mammograms”, in Proc. of the 3rd International Workshop on Digital Mammography (WDM), pp. 467–470. 2. Adam A., Omar, R. (2006.) “Computerized breast cancer diagnosis with Genetic Algorithm and Neural Network”, in Proc. of the 3 rd International Conference on Artificial Intelligence and Engineering Segmentation of Cancerous Mammography using MATLAB © 2022 Global Journals 1 Year 2022 52 Global Journal of Science Frontier Research Volume XXII Issue ersion I VI ( A )