scholarly journals Lossless Compression of Medical Images based on the Differential Probability of Images

2020 ◽  
Vol 14 ◽  
Keyword(s):  
Medical Image ◽  
Large Scale ◽  
Medical Images ◽  
Lossless Compression ◽  
Differential Method ◽  
Huffman Coding ◽  
Difference Matrix ◽  
Differential Probability ◽  
Compression Effect ◽  
The Difference

Lossless compression is crucial in the remote transmission of large-scale medical image and the retainment of complete medical diagnostic information. The lossless compression method of medical image based on differential probability of image is proposed in this study. The medical image with DICOM format was decorrelated by the differential method, and the difference matrix was optimally coded by the Huffman coding method to obtain the optimal compression effect. Experimental results obtained using the new method were compared with those using Lempel–Ziv–Welch, modified run–length encoding, and block–bit allocation methods to verify its effectiveness. For 2-D medical images, the lossless compression effect of the proposed method is the best when the object region is more than 20% of the image. For 3-D medical images, the proposed method has the highest compression ratio among the control methods. The proposed method can be directly used for lossless compression of DICOM images.

scholarly journals TransMed: Transformers Advance Multi-Modal Medical Image Classification

Diagnostics ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 1384
Author(s):  
Yin Dai ◽  
Yifan Gao ◽  
Fayu Liu
Keyword(s):  
Image Analysis ◽  
Image Classification ◽  
Long Range ◽  
Medical Image ◽  
Large Scale ◽  
Medical Images ◽  
Medical Image Analysis ◽  
Lesion Detection ◽  
Tumor Segmentation ◽  
Medical Image Classification

Over the past decade, convolutional neural networks (CNN) have shown very competitive performance in medical image analysis tasks, such as disease classification, tumor segmentation, and lesion detection. CNN has great advantages in extracting local features of images. However, due to the locality of convolution operation, it cannot deal with long-range relationships well. Recently, transformers have been applied to computer vision and achieved remarkable success in large-scale datasets. Compared with natural images, multi-modal medical images have explicit and important long-range dependencies, and effective multi-modal fusion strategies can greatly improve the performance of deep models. This prompts us to study transformer-based structures and apply them to multi-modal medical images. Existing transformer-based network architectures require large-scale datasets to achieve better performance. However, medical imaging datasets are relatively small, which makes it difficult to apply pure transformers to medical image analysis. Therefore, we propose TransMed for multi-modal medical image classification. TransMed combines the advantages of CNN and transformer to efficiently extract low-level features of images and establish long-range dependencies between modalities. We evaluated our model on two datasets, parotid gland tumors classification and knee injury classification. Combining our contributions, we achieve an improvement of 10.1% and 1.9% in average accuracy, respectively, outperforming other state-of-the-art CNN-based models. The results of the proposed method are promising and have tremendous potential to be applied to a large number of medical image analysis tasks. To our best knowledge, this is the first work to apply transformers to multi-modal medical image classification.

scholarly journals OFCS: Optimized Framework of Compressive Sensing for Medical Images in Bottleneck Network Condition

2018 ◽  
Vol 8 (5) ◽  
pp. 2829
Author(s):  
Lakshminarayana M ◽  
Mrinal Sarvagya
Keyword(s):  
Compressive Sensing ◽  
Medical Image ◽  
Medical Images ◽  
Region Of Interest ◽  
Lossless Compression ◽  
Cost Effective ◽  
Signal Quality ◽  
Effective Solution ◽  
The Cost

Compressive sensing is one of teh cost effective solution towards performing compression of heavier form of signals. We reviewed the existing research contribution towards compressive sensing to find that existing system doesnt offer any form of optimization for which reason the signal are superiorly compressed but at the cost of enough resources. Therefore, we introduce a framework that optimizes the performance of the compressive sensing by introducing 4 sequential algorithms for performing Random Sampling, Lossless Compression for region-of-interest, Compressive Sensing using transform-based scheme, and optimization. The contribution of proposed paper is a good balance between computational efficiency and quality of reconstructed medical image when transmitted over network with low channel capacity. The study outcome shows that proposed system offers maximum signal quality and lower algorithm processing time in contrast to existing compression techniuqes on medical images.

scholarly journals Integrated HNAS Network Model Based Lossless Compression with Data Hiding using Parity Check in Medical Images

2020 ◽  
Vol 9 (4) ◽  
pp. 2656-2662
Keyword(s):  
Network Model ◽  
Data Hiding ◽  
Compression Ratio ◽  
Reversible Data Hiding ◽  
Medical Image ◽  
Medical Images ◽  
Care Delivery ◽  
Lossless Compression ◽  
Parity Check ◽  
Compression Technique

A massive volume of medical data is generating through advanced medical image modalities. With advancements in telecommunications, Telemedicine, and Teleradiologyy have become the most common and viable methods for effective health care delivery around the globe. For sufficient storage, medical images should be compressed using lossless compression techniques. In this paper, we aim at developing a lossless compression technique to achieve a better compression ratio with reversible data hiding. The proposed work segments foreground and background area in medical images using semantic segmentation with the Hierarchical Neural Architecture Search (HNAS) Network model. After segmenting the medical image, confidential patient data is hidden in the foreground area using the parity check method. Following data hiding, lossless compression of foreground and background is done using Huffman and Lempel-Ziv-Welch methods. The performance of our proposed method has been compared with those obtained from standard lossless compression algorithms and existing reversible data hiding methods. This proposed method achieves better compression ratio and a hundred percent reversible when data extraction.

scholarly journals Layered Lossless Compression Method of Massive Fault Recording Data

2022 ◽  
Vol 16 ◽  
pp. 17-25
Author(s):  
Jinhong Di ◽  
Pengkun Yang ◽  
Chunyan Wang ◽  
Lichao Yan
Keyword(s):  
Large Scale ◽  
Lossless Compression ◽  
Search Method ◽  
Parallel Search ◽  
Compression Method ◽  
Compression Effect ◽  
Record Data ◽  
Update Strategy ◽  
Lzw Algorithm ◽  
Fault Record

In order to overcome the problems of large error and low precision in traditional power fault record data compression, a new layered lossless compression method for massive fault record data is proposed in this paper. The algorithm applies LZW (Lempel Ziv Welch) algorithm, analyzes the LZW algorithm and existing problems, and improves the LZW algorithm. Use the index value of the dictionary to replace the input string sequence, and dynamically add unknown strings to the dictionary. The parallel search method is to divide the dictionary into several small dictionaries with different bit widths to realize the parallel search of the dictionary. According to the compression and decompression of LZW, the optimal compression effect of LZW algorithm hardware is obtained. The multi tree structure of the improved LZW algorithm is used to construct the dictionary, and the multi character parallel search method is used to query the dictionary. The multi character parallel search method is used to query the dictionary globally. At the same time, the dictionary size and update strategy of LZW algorithm are analyzed, and the optimization parameters are designed to construct and update the dictionary. Through the calculation of lossless dictionary compression, the hierarchical lossless compression of large-scale fault record data is completed. Select the optimal parameters, design the dictionary size and update strategy, and complete the lossless compression of recorded data. The experimental results show that compared with the traditional compression method, under this compression method, the mean square error percentage is effectively reduced, and the compression error and compression rate are eliminated, so as to ensure the integrity of fault record data, achieve the compression effect in a short time, and achieve the expected goal.

Secure and Efficient Medical Image Transmission by New Tailored Visual Cryptography Scheme with LS Compressions

2015 ◽  
Vol 7 (1) ◽  
pp. 26-50 ◽  
Author(s):  
S. Manimurugan ◽  
C. Narmatha
Keyword(s):  
Medical Image ◽  
Medical Images ◽  
Visual Cryptography ◽  
Lossless Compression ◽  
Image Transmission ◽  
Vital Role ◽  
Compression Algorithms ◽  
Merging Process ◽  
Visual Cryptography Scheme ◽  
Short Algorithm

Exchanging a medical image via network from one place to another place or storing a medical image in a particular place in a secure manner has become a challenge. To overwhelm this, secure medical image Lossless Compression (LC) schemes have been proposed. The original input grayscale medical images are encrypted by Tailored Visual Cryptography Encryption Process (TVCE) which is a proposed encryption system. To generate these encrypted images, four types of processes are adopted which play a vital role. These processes are Splitting Process, Converting Process, Pixel Process and Merging process. The encrypted medical image is compressed by proposed compression algorithms, i.e Pixel Block Short algorithm (PBSA) and one conventional Lossless Compression (LC) algorithm has been adopted (JPEG 2000LS). The above two compression methods are used to separate compression for encrypted medical images. And also, decompressions have been done in a separate manner. The encrypted output image which is generated from decompression of the proposed compression algorithm, JPEG 2000LS are decrypted by the Tailored Visual Cryptography Decryption Process (TVCD). To decrypt the encrypted grayscale medical images, four types of processes are involved. These processes are Segregation Process, Inverse Pixel Process, 8-Bit into Decimal Conversion Process and Amalgamate Process. However, this paper is focused on the proposed visual cryptography only. From these processes, two original images have been reconstructed which are given by two compression algorithms. Ultimately, two combinations are compared with each other based on the various parameters. These techniques can be implemented in the field for storing and transmitting medical images in a secure manner. The Confidentiality, Integrity and Availability (CIA property) of a medical image have also been proved by the experimental results. In this paper we have focused on only proposed visual cryptography scheme.

scholarly journals Content clustering for MRI Image compression using PPAM

2018 ◽  
Vol 7 (1.7) ◽  
pp. 126
Author(s):  
Alex David S ◽  
Almas Begum ◽  
Ravikumar S
Keyword(s):  
Medical Imaging ◽  
Image Compression ◽  
Medical Image ◽  
Medical Images ◽  
Lossless Compression ◽  
Clustering Method ◽  
Image Content ◽  
Content Clustering ◽  
Mri Image

Image compression helps to save the utilization of memory, data while transferring the images between nodes. Compression is one of the key technique in medical image. Both lossy and lossless compressions where used based on the application. In case of medical imaging each and every components of pixel is very important hence its nature to chose lossless compression medical images. MRI images are compressed after processing. Here in this paper we have used PPMA method to compress the MRI image. For retrieval of the compressed image content clustering method used.

scholarly journals Improving Image Segmentation by DFS Algorithm

2020 ◽  
Vol 8 (5) ◽  
pp. 3505-3510
Keyword(s):  
Image Segmentation ◽  
Medical Image ◽  
Large Scale ◽  
Dynamic Range ◽  
Medical Images ◽  
Histogram Equalization ◽  
Quality Parameters ◽  
Medical Image Segmentation ◽  
Complete Survey ◽  
Medical Imagining

Medical imagining has proven to be a significant field for examining human tissues non-intrusively. One of the subset of Imaging is the Image segmentation where in an image is split into significant regions which being later used for classification and performing analysis. This process is quiet complex as it involves accurately detecting and removing the affected part of the image containing abnormal tissues which are later being used for analysis. Image segmentation employs numerous techniques and approaches. Though there exist several methods and techniques for image segmentation but all of them can’t be implemented on medical images. The existing paper put forwards a complete survey and review concerning the medical image segmentation models, techniques, algorithms along with the challenges faced with the involvement of contrast filtering and large scale image processing perspectives. The technique of Discrete Feature Segmentation (DFS) is adopted for extracting the attributes related to a medical image. For improvising the contrast of an image, the popular method of Histogram equalization is utilized that basically enlarges the dynamic range of intensity. A method is recommended for defining the parameters of the Contrast-Limited Adaptive Histogram Equalization (CLAHE) by utilizing entropy of image. The CLAHE method that projects intensity levels concerning the medical images is backed up by evidence from detection trials and anecdotal evidence. For classifying the diseases in medical image, the prime emphasis is on the FCM (Fuzzy C-Means (FCM) algorithm. Present research paper compares various techniques of image enhancement considering their quality parameters (PSNR, Mean, MSE, Entropy, SN, Variance and RMS).

Relationship between total plasma homocysteine and the risk of aneurysms – a meta-analysis

VASA ◽  
2020 ◽  
pp. 1-6
Author(s):  
Hanji Zhang ◽  
Dexin Yin ◽  
Yue Zhao ◽  
Yezhou Li ◽  
Dejiang Yao ◽  
...  
Keyword(s):  
Large Scale ◽  
Meta Analysis ◽  
Single Species ◽  
Total Plasma ◽  
Control Groups ◽  
Randomized Controlled ◽  
Randomized Controlled Studies ◽  
The Difference ◽  
The Relationship ◽  
Healthy Participants

Summary: Our meta-analysis focused on the relationship between homocysteine (Hcy) level and the incidence of aneurysms and looked at the relationship between smoking, hypertension and aneurysms. A systematic literature search of Pubmed, Web of Science, and Embase databases (up to March 31, 2020) resulted in the identification of 19 studies, including 2,629 aneurysm patients and 6,497 healthy participants. Combined analysis of the included studies showed that number of smoking, hypertension and hyperhomocysteinemia (HHcy) in aneurysm patients was higher than that in the control groups, and the total plasma Hcy level in aneurysm patients was also higher. These findings suggest that smoking, hypertension and HHcy may be risk factors for the development and progression of aneurysms. Although the heterogeneity of meta-analysis was significant, it was found that the heterogeneity might come from the difference between race and disease species through subgroup analysis. Large-scale randomized controlled studies of single species and single disease species are needed in the future to supplement the accuracy of the results.

Sign in / Sign up

Export Citation Format

Share Document