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

Segmentation of Cancerous Mammography using MATLAB Samuel Yemoh Tetteh-Abaku α , Calvin Kwesi Gafrey σ & Moses Jojo Eghan ρ Abstract- Breast cancer is one of the main causes of cancer death in women. Detection is efficiently performed by using digital mammograms. Small clusters of micro calcifications appearing as a collection of white spots on mammograms show an early warning of breast cancer. Early detection performed on X-ray mammography is the key to improving breast cancer diagnosis. To increase radiologists’ diagnostic performance, several computer-aided diagnosis (CAD) schemes have been developed to improve the detection of primary identification of this disease. In this research, an attempt is made to develop an adaptive K-means clustering algorithm for breast image segmentation to detect microcalcifications. The method was tested over several images of image databases taken from Mammocare, Ghana for cancer research and diagnosis. The algorithm works faster so that any radiologist can take a clear decision about the appearance of microcalcifications by visual inspection of digital mammograms and detection accuracy has also improved as compared to some existing works. I. I ntroduction reast cancer is a type of cancer with the highest incident rates in women. It has been one of the major causes of death among women since the last decades and it has become an emergency for the healthcare systems of countries. It is commonly classified into four stages according to the size of tumors and degree of cancer spread from the breast to other body parts and takes years to develop[5]. Mammography is an imaging study that uses X- rays to image the breast to look for cancer. There are two main types of mammography: film-screen mammography and digital mammography also called full-field digital mammography or FFDM. The technique for performing them is the same. What differs is whether the images take the form of photographic films or digital files recorded directly onto a computer. Mammography also has its limitations. It is less reliable on the dense breast of young women or women who underwent surgical intervention in the breast because glandular and scar tissues are as radiopaque as abnormalities[9]. Furthermore, there is low-dose X-Ray radiation. The estimated sensitivity of radiologists in breast cancer screening is only about 75%. Double reading has been suggested to be an effective approach to improve sensitivity. To improve the accuracy of interpretation, a variety of Computer Assisted Detection (CAD) techniques have been proposed. Interpretation of mammograms mainly involves two major processes: Computer-Aided Detection (CADe) and Computer-Aided Diagnosis (CADi)[20]. It would be valuable to develop a CAD algorithm using extracted features from the breast profile region; region of interest (ROI). This would reduce the number of biopsies in patients with benign disease and thus avoid patients’ physical and mental suffering, with a bonus of reducing healthcare costs. Initial detection of the cancerous mammogram helps in the early diagnosis of a disease a diseased person which can reduce death possibilities. Methods developed for the detection of the malignant region of the mammograms may not be able to provide results successfully[1,15]. Finding an accurate, robust and efficient breast profile segmentation remains a challenging problem in digital mammography. Hence mammography misses about 17% and up to 50% of breast cancers due to the subtle and unstable appearances of breast cancer in their early stages[8]. To overcome this limitation. It is necessary to develop an approach that can segment malignant regions properly. II. M aterials and M ethod a) Sample of images used Mammogram images were retrieved from the website of Mammocare Ghana. The mammogram images were acquired from Ghanaian patients. The images consist of left and right breast images of fatty, fatty-glandular and dense-glandular breasts, true positive and true negative breast images, false positive B 1 Year 2022 45 © 2022 Global Journals Global Journal of Science Frontier Research Volume XXII Issue ersion I VI ( A ) Author σ : e-mail: calvin.gafrey@stu.ucc.edu.gh A significant method that first detects the cancerous region and then segment the area covered by malignant tissues was proposed. In this paper, the focus was placed on detecting malignant tissues which represent higher intensity values compared to background information and other regions of the breast. However, in the case of some normal dense tissues having similar intensities to the tumor region, it is necessary to detect tumor region excluding those regions successfully. In this research work, an attempt was made to study the effect of L*a*b color space K- means clustering on colour image segmentation. Several general-purpose algorithms have been developed for image segmentation including detection followed by segmentation of mammogram images based on simple image processing techniques using the L*a*b colour space K-means clustering algorithm which provides good results in real-time[10].