Novel MOS prediction models for compressed medical image quality

This paper presents the development of novel models which can be potentially useful in determining the upper limit of image compression thresholds, to preserve diagnostically relevant information in compressed medical images. These models were developed by evolving the correlation between the theoretically computed objective (peak signal-to-noise ratio and structural similarity) and subjective mean opinion score (MOS) quality parameters. The developed models were validated by comparing the model generated MOS with the corresponding experimental MOS of six independent observers considering joint photographic experts group (JPEG), JPEG2000 and set partitioning in hierarchical trees (SPIHT) compressions of computed tomography (CT) scan images. It is found that the correlation between the model generated and experimental MOS and PRD are ≥0.87 and ≤13% respectively for the compression range 0.05-2.0 bits/pixel of the CT scan images. Therefore our models can be potentially useful for observer-independent MOS prediction and quality assessment of reconstructed medical images. In addition this also avoids the need for exhaustive and time-consuming experimental MOS and thus it can be more suitable for teleradiology applications.     

基于离散小波变换的JPEG 2000标准对肺部CT图像的无损压缩

探索基于小波变换对肺部CT图像进行无损雎缩的新方法.采用基于离散小波变换方法的JPEG2000标准对3019张肺部CT图像进行无损压缩,并针对压缩后图像效果进行统计分析.结果表明,该方法不但可以达到12.0的高压缩比,而且具有较高的图像尤损压缩质量,为临床CT医学影像的储存与诊断,提供了有益的技术与方法.     

A Mobile Tele-Radiology Imaging System with JPEG2000 for an Emergency Care

The aim of this study was to design a tele-radiology imaging system for rapid emergency care via mobile networks and to assess the diagnostic feasibility of the Joint Photographic Experts Group 2000 (JPEG2000) radiological imaging using portable devices. Rapid patient information and image exchange is helpful to make clinical decisions. We assessed the usefulness of the mobile tele-radiology system by measuring both a quantitative method, PNSR calculation, for image qualities, and its transmission time via mobile networks in different mobile networks, respectively; code division multiple access evolution-data optimized, wireless broadband, and high-speed downlink packet access; and the feasibility of the JPEG2000 computed tomography (CT) images by qualitatively assessing with the Alberta stroke program early CT score method with 12 CT image cases (seven normal and five abnormal cases). We found that the quality of the JPEG2000 radiological images was satisfied quantitatively and was judged as acceptable qualitatively at 5:1 and 10:1 compression levels for the mobile tele-radiology imaging system. The JPEG2000-format radiological images achieved a fast transmission while maintaining a diagnosis quality on a portable device via mobile networks. Unfortunately, a PDA device, having a limited screen resolution, posed difficulties in reviewing the JPEG2000 images regardless of the compression levels. An ultra mobile PC was preferable to study the medical image. The mobile tele-radiology imaging systems supporting JPEG2000 image transmission can be applied to actual emergency care services under mobile computing environments.     

Pan-Canadian Evaluation of Irreversible Compression Ratios (“Lossy” Compression) for Development of National Guidelines

New technological advancements including multislice CT scanners and functional MRI, have dramatically increased the size and number of digital images generated by medical imaging departments. Despite the fact that the cost of storage is dropping, the savings are largely surpassed by the increasing volume of data being generated. While local area network bandwidth within a hospital is adequate for timely access to imaging data, efficiently moving the data between institutions requires wide area network bandwidth, which has a limited availability at a national level. A solution to address those issues is the use of lossy compression as long as there is no loss of relevant information. The goal of this study was to determine levels at which lossy compression can be confidently used in diagnostic imaging applications. In order to provide a fair assessment of existing compression tools, we tested and compared the two most commonly adopted DISCOM compression algorithms: JPEG and JPEG-2000. We conducted an extensive pan-Canadian evaluation of lossy compression applied to seven anatomical areas and five modalities using two recognized techniques: objective methods or diagnostic accuracy and subjective assessment based on Just Noticeable Difference. By incorporating both diagnostic accuracy and subjective evaluation techniques, enabled us to define a range of compression for each modality and body part tested. The results of our study suggest that at low levels of compression, there was no significant difference between the performance of lossy JPEG and lossy JPEG 2000, and that they are both appropriate to use for reporting on medical images. At higher levels, lossy JPEG proved to be more effective than JPEG 2000 in some cases, mainly neuro CT. More evaluation is required to assess the effect of compression on thin slice CT. We provide a table of recommended compression ratios for each modality and anatomical area investigated, to be integrated in the Canadian Association of Radiologists standard for the use of lossy compression in medical imaging.     

Clinical Evaluation of Compression Ratios using JPEG2000 on Computed Radiography Chest Images

The efficient compression of radiographic images is of importance for improved storage and network utilization in support of picture archiving and communication systems (PACS) applications. The DICOM Working Group 4 adopted JPEG2000 as an additional compression standard in Supplement 61 over the existing JPEG. The wavelet-based JPEG2000 can achieve higher compression ratios with less distortion than the Discrete Cosine Transform (DCT)-based JPEG algorithm. However, the degradation of JPEG2000-compressed computed radiography (CR) chest images has not been tested comprehensively clinically. The authors evaluated the diagnostic quality of JPEG2000-compressed CR chest images with compression ratios from 5:1 to 200:1. An ROC (receiver operating characteristic analysis) and t test were performed to ascertain clinical performance using the JPEG2000-compressed images. The authors found that compression ratios as high as 20:1 can be utilized without affecting lesion detectability. Significant differences between the original and the compressed CR images were not recognized up to compression ratio of 50:1 within a confidence level of 99%.     

Quality of Compressed Medical Images

Previous studies have shown that Joint Photographic Experts Group (JPEG) 2000 compression is better than JPEG at higher compression ratio levels. However, some findings revealed that this is not valid at lower levels. In this study, the qualities of compressed medical images in these ratio areas (∼20), including computed radiography, computed tomography head and body, mammographic, and magnetic resonance T1 and T2 images, were estimated using both a pixel-based (peak signal to noise ratio) and two 8 × 8 window-based [Q index and Moran peak ratio (MPR)] metrics. To diminish the effects of blocking artifacts from JPEG, jump windows were used in both window-based metrics. Comparing the image quality indices between jump and sliding windows, the results showed that blocking artifacts were produced from JPEG compression, even at low compression ratios. However, even after the blocking artifacts were omitted in JPEG compressed images, JPEG2000 outperformed JPEG at low compression levels. We found in this study that the image contrast and the average gray level play important roles in image compression and quality evaluation. There were drawbacks in all metrics that we used. In the future, the image gray level and contrast effect should be considered in developing new objective metrics.     

A comparison of wavelet and Joint Photographic Experts Group lossy compression methods applied to medical images

This presentation focuses on the quantitative comparison of three lossy compression methods applied to a variety of 12-bit medical images. One Joint Photographic Exports Group (JPEG) and two wavelet algorithms were used on a population of 60 images. The medical images were obtained in Digital Imaging and Communications in Medicine (DICOM) file format and ranged in matrix size from 256 × 256 (magnetic resonance [MR]) to 2,560 × 2,048 (computed radiography [CR], digital radiography [DR], etc). The algorithms were applied to each image at multiple levels of compression such that comparable compressed file sizes were obtained at each level. Each compressed image was then decompressed and quantitative analysis was performed to compare each compressed-thendecompressed image with its corresponding original image. The statistical measures computed were sum of absolute differences, sum of squared differences, and peak signal-to-noise ratio (PSNR). Our results verify other research studies which show that wavelet compression yields better compression quality at constant compressed file sizes compared with JPEG. The DICOM standard does not yet include wavelet as a recognized lossy compression standard. For implementers and users to adopt wavelet technology as part of their image management and communication installations, there has to be significant differences in quality and compressibility compared with JPEG to justify expensive software licenses and the introduction of proprietary elements in the standard. Our study shows that different wavelet implementations vary in their capacity to differentiate themselves from the old, established lossy JPEG.     

Adaptive threshold-based block classification in medical image compression for teleradiology

Telemedicine, among other things, involves storage and transmission of medical images, popularly known as teleradiology. Due to constraints on bandwidth and storage capacity, a medical image may be needed to be compressed before transmission/storage. Among various compression techniques, transform-based techniques that convert an image in spatial domain into the data in spectral domain are very effective. Discrete cosine transform (DCT) is possibly the most popular transform used in compression of images in standards like Joint Photographic Experts Group (JPEG). In DCT-based compression the image is split into smaller blocks for computational simplicity. The blocks are classified on the basis of information content to maximize compression ratio without sacrificing diagnostic information. The present paper presents a technique along with computational algorithm for classification of blocks on the basis of an adaptive threshold value of variance. The adaptive approach makes the classification technique applicable across the board to all medical images. Its efficacy is demonstrated by applying it to CT, X-ray and ultrasound images and by comparing the results against the JPEG in terms of various objective quality indices.     

The impact of image information on compressibility and degradation in medical image compression

The aim of the study was to demonstrate and critically discuss the influence of image information on compressibility and image degradation. The influence of image information on image compression was demonstrated on the axial computed tomography images of a head. The standard Joint Photographic Expert Group (JPEG) and JPEG 2000 compression methods were used in compression ratio (CR) and in quality factor (QF) compression modes. Image information was estimated by calculating image entropy, while the effects of image compression were evaluated quantitatively, by file size reduction and by local and global mean square error (MSE), and qualitatively, by visual perception of distortion in high and low contrast test patterns. In QF compression mode, a strong correlation between image entropy and file size was found for JPEG (r=0.87, p < 0.001) and JPEG 2000 (r=0.84, p < 0.001), while corresponding local MSE was constant (4.54) or nearly constant (2.36-2.37), respectively. For JPEG 2000 CR compression mode, CR was nearly constant (1:25), while local MSE varied considerably (2.26 and 10.09). The obtained qualitative and quantitative results clearly demonstrate that image degradation highly depends on image information, which indicates that the degree of image degradation cannot be guaranteed in CR but only in QF compression mode. CR is therefore not a measure of choice for expressing the degree of image degradation in medical image compression. Moreover, even when using QF compression modes, objective evaluation, and comparison of the compression methods within and between studies is often not possible due to the lack of standardization of compression quality scales.     

JPEG2000 Still Image Coding Quality

This work demonstrates the image qualities between two popular JPEG2000 programs. Two medical image compression algorithms are both coded using JPEG2000, but they are different regarding the interface, convenience, speed of computation, and their characteristic options influenced by the encoder, quantization, tiling, etc. The differences in image quality and compression ratio are also affected by the modality and compression algorithm implementation. Do they provide the same quality? The qualities of compressed medical images from two image compression programs named Apollo and JJ2000 were evaluated extensively using objective metrics. These algorithms were applied to three medical image modalities at various compression ratios ranging from 10:1 to 100:1. Following that, the quality of the reconstructed images was evaluated using five objective metrics. The Spearman rank correlation coefficients were measured under every metric in the two programs. We found that JJ2000 and Apollo exhibited indistinguishable image quality for all images evaluated using the above five metrics (r > 0.98, p < 0.001). It can be concluded that the image quality of the JJ2000 and Apollo algorithms is statistically equivalent for medical image compression.     

医用内窥镜图像自适应量化压缩编化码方法研究

本文提出了一种与ＪＰＥＧ标准完全兼容的医用内窥镜图像自适应量化压缩编码方法,方法采用二次扫描的措施,根据原始图像的频谱分布特点,自适应地修正ＪＰＥＧ推荐的量化表。实验结果表明：该方法较之于标准ＪＰＥＧ图像压缩,峰值信噪比（ＰＳＮＲ）明显提高,可在相同压缩比下,保持更多的图像细节,特别适合于医学图像的压缩。     

Medical ultrasound image compression using contextual vector quantization

With ever increasing use of medical ultrasound (US) images, a challenge exists to deal with storage and transmission of these images while still maintaining high diagnostic quality. In this article, a state-of-the-art context based method is proposed to overcome this challenge called contextual vector quantization (CVQ). In this method, a contextual region is defined as a region containing the most important information and must be encoded without considerable quality loss. Attempts are made to encode this region with high priority and high resolution (low compression ratio and high bit rate) CVQ algorithm; and the background, which has a lower priority, is separately encoded with a low resolution (high compression ratio and low bit rate) version of the CVQ algorithm. Finally both of the encoded contextual region and the encoded background region is merged together to reconstruct the output image. As a result, very good diagnostic image quality with lower image size and enhanced performance parameters including mean square error (MSE), pick signal to noise ratio (PSNR) and coefficient of correlation (CoC) are gained. The experimental results show that the proposed CVQ methodology is superior as compared to other existing methods (general methods such as JPEG and JPEG2K, and ROI based methods such as EBCOT and CSPIHT) in terms of measured performance parameters. This makes CVQ compression method a feasible technique to overcome storage and transmission limitations.     

Evaluation of Irreversible Compression Ratios for Medical Images Thin Slice CT and Update of Canadian Association of Radiologists (CAR) Guidelines

In June 2008, the Canadian Association of Radiologists published its Standards for Irreversible Compression in Digital Diagnostic Imaging within Radiology (Canadian Association of Radiologists 2012). The study suggested that at low levels of compression there was no difference in diagnostic accuracy between uncompressed JPEG and JPEG 2000. There were two exceptions; CT neurological and CT body images resulted in lower rating of image quality (Koff et al., J Digit Imaging 22(6):569–78, 2009). The slice thicknesses used in the previous study were greater than 5 mm. However, other studies (Ringl et al., Radiology 240:869–87, 2006) suggest that thin CT slices might modify image tolerance to irreversible compression. Therefore, a new clinical evaluation using CT slices less than 3 mm was initiated. We examined CT images in four body regions (chest, body, musculoskeletal, and neurological). Twenty-five radiologists from across Canada participated. Each read a total of 70 CTs in his specialty; 10 at each of seven levels of compression (uncompressed, JPEG and JPEG 2000 at low, medium, and high compression (varying by region)). Each reader diagnosed the case, rated his confidence, and compared the compressed to the uncompressed image and rated the degree of degradation. Data were analyzed for sensitivity, specificity, accuracy, confidence, and degradation at three levels and two types of compression as well as the original image. There were no overall differences in sensitivity, specificity, accuracy, or confidence. JPEG images, at all levels of compression, were rated lower in terms of perceived difference (4.16/5 vs. 4.53/5 for JPEG 2000 and 4.68/5 for uncompressed). However, the rating of perceived difference was not significantly correlated with accuracy. Analysis of individual body regions did not reveal any systematic effects of compression in any region.     

Algorithms for density and composition-discrimination imaging for fourth-generation CT systems.

This paper shows that if the off-beam idle detectors in the detection ring of a fourth-generation x-ray computed tomography (CT) system are used to measure the scattered radiation, it is numerically feasible to reconstruct electron-density images to supplement the conventional attenuation-coefficient images of transmitted radiation. It is also shown that by combining these two images, composition changes can be detected with the aid of an effective-atomic-number indicator. The required image-reconstruction algorithms are developed and tested against Monte Carlo simulated measurements, for a variety of phantom configurations. In spite of the relatively poor statistical quality of scattering measurements, it is demonstrated that electron-density images of reasonable quality can be obtained. In addition, it is shown that composition discrimination is possible for materials of effective atomic number greater than five, in the photon energy range of a typical medical x-ray CT system operating at 102 kVp. The obtained supplementary electron-density and composition images can be useful in radiotherapy planning and for studying tumour histology, as well as in industrial and security applications where identification of materials based on density and composition is important.     

Trabecular Level Analysis of Bone Cement Augmentation: A Comparative Experimental and Finite Element Study

The representation of cement-augmented bone in finite element (FE) models of vertebrae following vertebroplasty remains a challenge, and the methods of the model validation are limited. The aim of this study was to create specimen-specific FE models of cement-augmented synthetic bone at the microscopic level, and to develop a new methodology to validate these models. An open cell polyurethane foam was used reduce drying effects and because of its similar structure to osteoporotic trabecular bone. Cylindrical specimens of the foam were augmented with PMMA cement. Each specimen was loaded to three levels of compression inside a micro-computed tomography (μCT) scanner and imaged both before compression and in each of the loaded states. Micro-FE models were generated from the unloaded μCT images and displacements applied to match measurements taken from the images. A morphological comparison between the FE-predicted trabecular deformations and the corresponding experimental measurements was developed to validate the accuracy of the FE model. The predicted deformation was found to be accurate (less than 12% error) in the elastic region. This method can now be used to evaluate real bone and different types of bone cements for different clinical situations.     

JPEG2000 for automated quantification of immunohistochemically stained cell nuclei: a comparative study with standard JPEG format

The Joint Photographic Experts Group (JPEG) standard format is one of the most widely used in image compression technologies. More recently, JPEG2000 format has emerged as a state-of-the-art technology that provides substantial improvements in picture quality at higher compression ratios. However, there has been no attempt to date to determine which of the two compression formats produces less variability in the automated evaluation of immunohistochemically stained digital images in agreement with their compression rates and complexity degrees. The evaluation of Ki67 and FOXP3 immunohistochemical nuclear markers was performed in a total of 329 digital images: 47 were captured in uncompressed Tagged Image File Format (TIFF), 141 were converted to three JPEG compressed formats (47 each with 1:3, 1:23 and 1:46 compression) and 141 were converted to three JPEG2000 compressed formats (47 each with 1:3, 1:23 and 1:46 compression). The count differences between images in TIFF versus JPEG formats were compared with those obtained between images in TIFF versus JPEG2000 formats at the three levels of compression. It was found that, using JPEG2000 compression, the results of the stained nuclei count are close enough to the results obtained with uncompressed images, especially in highly complex images at minimum and medium compression. Otherwise, in images of low complexity, JPEG and JPEG2000 had similar count efficiency to that of the original TIFF images at all compression levels. These data suggest that JPEG2000 could give rise to an efficient means of storage, reducing file size and storage capacity, without compromise on the immunohistochemical analytical quality.     

Performance characterization of megavoltage computed tomography imaging on a helical tomotherapy unit

Helical tomotherapy is an innovative means of delivering IGRT and IMRT using a device that combines features of a linear accelerator and a helical computed tomography (CT) scanner. The HI-ART II can generate CT images from the same megavoltage x-ray beam it uses for treatment. These megavoltage CT (MVCT) images offer verification of the patient position prior to and potentially during radiation therapy. Since the unit uses the actual treatment beam as the x-ray source for image acquisition, no surrogate telemetry systems are required to register image space to treatment space. The disadvantage to using the treatment beam for imaging, however, is that the physics of radiation interactions in the megavoltage energy range may force compromises between the dose delivered and the image quality in comparison to diagnostic CT scanners. The performance of the system is therefore characterized in terms of objective measures of noise, uniformity, contrast, and spatial resolution as a function of the dose delivered by the MVCT beam. The uniformity and spatial resolutions of MVCT images generated by the HI-ART II are comparable to that of diagnostic CT images. Furthermore, the MVCT scan contrast is linear with respect to the electron density of material imaged. MVCT images do not have the same performance characteristics as state-of-the art diagnostic CT scanners when one objectively examines noise and low-contrast resolution. These inferior results may be explained, at least partially, by the low doses delivered by our unit; the dose is 1.1 cGy in a 20 cm diameter cylindrical phantom. In spite of the poorer low-contrast resolution, these relatively low-dose MVCT scans provide sufficient contrast to delineate many soft-tissue structures. Hence, these images are useful not only for verifying the patient's position at the time of therapy, but they are also sufficient for delineating many anatomic structures. In conjunction with the ability to recalculate radiotherapy doses on these images, this enables dose guidance as well as image guidance of radiotherapy treatments.     

Technical note: acquisition of CT models for radiotherapy applications with reduced tube heating

The potential or changing computed tomography (CT) protocols to provide data sets that generate high quality digitally reconstructed radiographs (DRRs) from scans with very low tube currents is demonstrated. DRRs were generated from CT data acquired with slice thickness of 1, 3, and 5 mm, using high current to reduce noise in axial images. These DRRs were compared to one generated from a CT scan acquired using 1 mm aperture and very low (10 mA) current. The DRR generated via this technique is comparable to that generated with high current and 1 mm aperture, and higher resolution than from the 3 and 5 mm CT scans.     

Automated Facial Recognition of Computed Tomography-Derived Facial Images: Patient Privacy Implications

The recognizability of facial images extracted from publically available medical scans raises patient privacy concerns. This study examined how accurately facial images extracted from computed tomography (CT) scans are objectively matched with corresponding photographs of the scanned individuals. The test subjects were 128 adult Americans ranging in age from 18 to 60 years, representing both sexes and three self-identified population (ancestral descent) groups (African, European, and Hispanic). Using facial recognition software, the 2D images of the extracted facial models were compared for matches against five differently sized photo galleries. Depending on the scanning protocol and gallery size, in 6–61 % of the cases, a correct life photo match for a CT-derived facial image was the top ranked image in the generated candidate lists, even when blind searching in excess of 100,000 images. In 31–91 % of the cases, a correct match was located within the top 50 images. Few significant differences (p > 0.05) in match rates were observed between the sexes or across the three age cohorts. Highly significant differences (p < 0.01) were, however, observed across the three ancestral cohorts and between the two CT scanning protocols. Results suggest that the probability of a match between a facial image extracted from a medical scan and a photograph of the individual is moderately high. The facial image data inherent in commonly employed medical imaging modalities may need to consider a potentially identifiable form of “comparable” facial imagery and protected as such under patient privacy legislation.     

联合图像专家组2000图像压缩方法的核医学应用研究

为研究联合图像专家组2000(Jo in t Photograph ic Expert G roup 2000,JPEG 2000)图像压缩方法在核医学中的应用,将无病变和有病变核医学静态图像用JPEG 2000软件压缩。对无损压缩图像,测量其压缩比。对有损压缩图像,由医生阅片,根据其结论作接收器操作特性(R ece iver operating characteristic,ROC)分析,获得各种图像压缩比的ROC曲线下的面积(A rea under curve,AUC),以其大小评价图像诊断质量;并将原始图像组AUC与各有损压缩图像组的AUC作配对t检验。实验发现,无损压缩的图像压缩比为(1.34±0.05)∶1。而有损压缩比越大,AUC越小。原始图像与压缩图像比较,压缩比为10∶1时没有显著性差异,压缩比更大时则有显著性差异。实验结果表明,无损压缩方法压缩比低,实用意义不大。有损压缩比不大于10∶1时,核医学静态图像的诊断质量得以保留。对核医学中的其它图像形式,可根据的图像性质,特别是固有统计噪声的大小,适当增减压缩比。