Channel Transformation Enhanced Deep Convolutional Neural Network enforced Image Retrieval Mechanism for Medical Image Applications
Senthil Kumar Sundararajan1, B. Sankaragomathi2, D. Saravana Priya3
1Senthil Kumar Sundararajan, Research Scholar, Department of Computer Science, Bharathiar University, Coimbatore (Tamil Nadu), India.
2Dr. B. Sankaragomathi, Professor, Department of Electronics and Instrumentation, National Engineering College, Kovilpatti (Tamil Nadu), India.
3Dr. D. Saravana Priya, Department of Computer Science Engineering, PA College of Engineering and Technology, Pollachi (Tamil Nadu), India.
Manuscript received on 24 April 2019 | Revised Manuscript received on 02 May 2019 | Manuscript Published on 08 May 2019 | PP: 405-411 | Volume-7 Issue-5S3 February 2019 | Retrieval Number: E11730275S19/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The diversified utilization of digital imaging data in the medical domain has in turn increased the size of the medical image repository. This increase in the size of the repository imposes huge challenges during the process of querying and handling huge databases will lead to the requirement of Content Based Medical Image Retrieval Systems(CBMIR). In this paper, a Channel Transformation Enhanced Deep Convolutional Neural Network-based Image Retrieval Mechanism (CTEDCNN-IRM) is proposed for handling the issue of semantic gap that prevails between human perceived high level semantic information and imaging devices’ captured low level visual information in medical imaging applications. The experimental results of the proposed CTEDCNN-IRM confirmed a mean classification accuracy and mean precision rate of 99.83% and 0.78 in the process of image retrieval. This proposed CTEDCNNIRM is also determined to be well suited and applicable to the processing of multimodal medical images that relates to different body organs.
Keywords: Transformation Neural Network Image Retrieval Applications Information Process.
Scope of the Article: Information Retrieval