<title>A novel content-based medical image retrieval method based on query topic dependent image features (QTDIF)</title>

Today, the common problem in the domain of computer vision and pattern recognition is content based image retrieval (CBIR). In this paper, a novel image retrieval method using the geometric details based on the correlation among edgels and correlation between pixels has been introduced. The autocorrelation based choridiogram descriptor has been extracted from the image to obtain geometric, texture and spatial information. Color autocorrelogram has been computed to obtain color, texture and spatial information. The proposed method is tested on benchmark heterogeneous medical image database and LIDC-IDRI-CT and VIA/I-ELCAP-CT databases and results are compared with typical CBIR system for medical image retrieval

Download Full-text

An Enhanced Framework for Content Based Medical Image Retrieval using Deep Neural Network

International Journal of Engineering and Advanced Technology - Regular Issue ◽

10.35940/ijeat.e1054.0785s319 ◽

2019 ◽

Vol 8 (5S3) ◽

pp. 239-244

Keyword(s):

Image Retrieval ◽

Medical Image ◽

Low Cost ◽

Image Features ◽

Semantic Gap ◽

Medical Image Retrieval ◽

Large Size ◽

Image Repository ◽

High Level ◽

Technology Growth

As the technology growth fuelled by low cost tech in the areas of compute, storage the need for faster retrieval and processing of data is becoming paramount for organizations. The medical domain predominantly for medical image processing with large size is critical for making life critical decisions. Healthcare community relies upon technologies for faster and accurate retrieval of images. Traditional, existing problem of efficient and similar medical image retrieval from huge image repository are reduced by Content Based Image Retrieval (CBIR) . The major challenging is an semantic gap in CBIR system among low and high level image features. This paper proposed, enhanced framework for content based medical image retrieval using DNN to overcome the semantic gap problem. It is outlines the steps which can be leveraged to search the historic medical image repository with the help of image features to retrieve closely relevant historic image for faster decision making from huge volume of database. The proposed system is assessed by inquisitive amount of images and the performance efficiency is calculated by precision and recall evaluation metrics. Experimental results obtained the retrieval accuracy is 79% based on precision and recall and this approach is preformed very effectively for image retrieval performance.

Download Full-text

Convolutional Neural Networks in Image Retrieval System

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.d2001.039520 ◽

2020 ◽

Vol 9 (5) ◽

pp. 123-129

Keyword(s):

Neural Networks ◽

Deep Learning ◽

Image Retrieval ◽

Convolutional Neural Networks ◽

Medical Image ◽

Retrieval System ◽

Prediction Models ◽

Image Features ◽

Medical Image Retrieval ◽

Image Retrieval System

Image is an important medium for monitoring the treatment responses of patient’s diseases by the physicians. There could be a tough task to organize and retrieve images in structured manner with respect to incredible increase of images in Hospitals. Text based image retrieval may prone to human error and may have large deviation across different images. Content-Based Medical Image Retrieval(CBMIR) system plays a major role to retrieve the required images from the huge database.Recent advances in Deep Learning (DL) have made greater achievements for solving complex problems in computer vision ,graphics and image processing. The deep architecture of Convolutional Neural Networks (CNN) can combine the low-level features into high-level features which could learn the semantic representation from images. Deep learning can help to extract, select and classify image features, measure the predictive target and gives prediction models to assist physician efficiently. The motivation of this paper is to provide the analysis of medical image retrieval system using CNN algorithm.

Download Full-text

Medical Image Retrieval Based on Simplified Multi-Wavelet Transform and Shape Feature

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.513-517.2871 ◽

2014 ◽

Vol 513-517 ◽

pp. 2871-2875

Author(s):

Xin Rui Wang ◽

Yun Feng Yang

Keyword(s):

Wavelet Transform ◽

Image Retrieval ◽

High Frequency ◽

Medical Image ◽

Shape Feature ◽

Retrieval Strategy ◽

Canny Operator ◽

Retrieval Method ◽

Medical Image Retrieval ◽

Contour Lines

A novel medical image retrieval method based on Simplified Multi-wavelet Transform and Shape Feature was proposed in the paper, which included coarse and fine retrieval procedure. In the procedure of the coarse retrieval, Canny operator was used to extract edges of images. Moreover, contour lines were obtained by using the method of scan lines. At last, the coarse retrieval results of the images can be accomplished by using contour lines of images. In the procedure of the fine retrieval, the simplified multi-wavelet transform was used to decompose images at first, then, only the high frequency coefficients in the vertical directions were selected as retrieval objects. And hierarchical retrieval strategy was selected to accomplish the fine retrieval. This method not only can reduce the computational complexity effectively, but also can make full use of high frequency information of original images. Experiments showed that the accuracy of the retrieved results can be ensured.

Download Full-text