Detection of Tumors and Stages of Mammae Carcinoma on Mammography Imaging Using Morphological Operations
Year: 2026
Type: article
Abstract: Carcinoma of the mammae is disease caused the changing breast cells and forming malignant tumors. Mammography plays an important role as an initial screening method, while histopathological results remain the gold standard for determining the stage of mammary carcinoma. And in the previous method, thresholding limitations did not separate noise, besides that Region Limitation Contrast technique was not good and Active Contour time was relatively long, therefore, Machine learning K-means clustering and Morphological operations became an accurate diagnosis solution through image algorithms, Machine learning have the ability to represent features that have similarities in expert results with Radiologists and even histopathology Results. Able to classify radiography results automatically detect the stage ca mammae and have the same results histopathology. Quasi experimental research on Post-test Only Control Group Design. Build machine learning Matlab program. The test measures accuracy, sensitivity, specificity, positive prediction value and NPV. Data analysis validity test and Wilcoxon statistical test. The study proved that 164 samples obtained good machine learning performance in detecting the stage mammary carcinoma with an accuracy value 97.35%, sensitivity 85.74%, specificity 96.92%, positive prediction value 88.04%, NPV 96.91%, there was similarity between Machine learning and histopathology results.
Source: Jurnal Penelitian Pendidikan IPA
Institution: Politeknik Muhammadiyah Makassar
Language: English
Topic: AI in cancer detection
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7204253769