Comparative study on the efficiency of digital tomosynthesis and conventional mammogram pertains to diseased BI-Raids 3 lesions using various classifiers for potential cancer diagnosis Available to Purchase

The goal of employing image segmentation in breast imaging is to improve upon regular mammograms and allow for more accurate detection of cancerous areas in calcifications. Two databases containing information on breast cancer diseases were used to extract 48 samples in total. We were able to assess the digital tom synthesis and conventional mammography image segmentation of BI-Raid lesion diseased breast cancer by comparing the cancer spot detection rates and computing the sensitivity and accuracy values. We used MATLAB R2018a to build the photo segmentation system. The parameters that were used to compute the sample size were: a confidence interval of 95%, α=0.05 and G-power set at 80%. Compared to traditional mammography methods, the research found that K-NN classifier performed much better in detecting cancer spots in digital tom synthesis images with BI-Raid lesions. When testing on pictures from digital tom synthesis and traditional mammography for breast cancer, the K-NN classifier achieved an accuracy rate of 79.54%. Previous research indicated that the K-NN classifier was more accurate than the SVM and K-NB classifiers, which had accuracy rates of 78.15% and 78.01%, respectively. The K-NN classifier was found to have a significance value of P<0.002. Digital tom synthesis pictures of breast cancer showed better picture segmentation accuracy than traditional mammography images when employing a K-NN classifier, according to the study’s independent testing. Hence, compared to traditional mammography, the K-NN classifier performs better at identifying cancerous regions in breast cancer digital tom synthesis pictures.