الاكتشاف المبكر لسرطان الرحم باستخدام تقنية معالجة الصورة
الخلاصة
According to GLOBOCAN 2020 estimates of cancer incidence and mortality, Endometrium cancer is the second leading cause of mortality in women after Breast cancer. However, it is also one of the treatable cancers if detected early. Radiologists read uterine ultrasound images manually, which most of the time, is a relatively difficult and confusing procedure that causes them to make mistakes. The focus of this research was to look at the possibilities of detecting and classifying Endometrium cancer using image-processing techniques. The study used filtering techniques to enhance the image enhancement process and used fuzzy logic to enhance edge detection of the field of interest. To improve the image detection process, the quality of the input ultrasound image was first improved during the pre-processing stage by removing noise using a median filter. Edges were then detected using fuzzy logic. Two techniques were then used to obtain the region of interest for the ultrasound image of the uterus, which is the endometrium region. The first technique is k-means, and the second is automatic thresholding. The researcher used two methods to evaluate the experiment. The first one involved measuring the quality of the resulting image using quality assessment equations which are Peak Signal to Noise, Mean squire Errors, and Ratio Mean Absolute Errors. The second method involved conducting a questionnaire to evaluate the perceived quality of the processed image after three stages: edge detection, applying the k-means technique, and using automatic thresholding. The researcher also sought the expertise of doctors from the Department of Obstetrics and Gynecology at the National Cancer Institute in Sabratha to determine the best image that clearly shows the uterine lining. The results of the automatic thresholding technique were well-received by nine out of ten doctors. Based on this, recommendations were made to improve the experiment. These included using a large database to test the success of this experiment and utilizing other features that doctors rely on to detect cancer, such as the thickness of the uterine lining and the widening of the uterine cavity. This would aid in early detection and ultimately save patients' lives.