Medical image classification is an effective and efficient technique for retrieving images from a large dataset containing similar images. A robust recovery system requires an accurate and accurate classification of information in the form of shape, composition and colors. In this work, a deep convolutional Intensive Pragmatic Blossoms (IPB) technique is proposed to classify the information in the collection of medical images. The IPB need millions of data, but the lack of large uniform data availability that the medical domain encourages us to believe is even better than the second prediction type. This will result in a proper classification performance. Furthermore, radon conversion is popular in the field of medicine, and a similarity-based search program is using this transformation technique. The test results and the comparison show that this strategy not only improves performance but also can save on computational costs. The simulation results identify the suggested methodology and its effectiveness in classifying the images of synthesized medicine. Hence the proposed model produces better features for clustering in medical images. As Compared to another conventional classifier, our proposed IPB method performance by achieving an accuracy of 98.03%,precision of 93.70%,recall of 93.70 % and F-Measure of 5.47%
Volume 12 | Issue 5
Pages: 216-230
DOI: 10.5373/JARDCS/V12I5/20201708