Tzong-Jer Chen,1Keh-Shih Chuang, Ph.D., Jay Wu,1,2Sharon C. Chen,1 Ing-Ming Hwang,1,3and Meei-Ling Jan1,4
The objective of this study was to develop a method for measuring quality degradation in lossy wavelet image compression. Quality degradation is due to denoising and edge blurring effects that cause smoothness in the compressed image. The peak Moran z histogram ratio between the reconstructed and original images is used as an index for degrada-tion after image compression. The Moran test is ap-plied to images randomly selected from each medical modality, computerized tomography, magnetic reso-nance imaging, and computed radiography and compressed using the wavelet compression at vari-ous levels. The relationship between the quality degradation and compression ratio for each image modality agrees with previous reports that showed a preference for mildly compressed images. Preliminary results show that the peak Moran z histogram ratio can be used to quantify the quality degradation in lossy image compression. The potential for this method is applications for determining the optimal compression ratio (the maximized compression without seriously degrading image quality) of an im-age for teleradiology.
KEY WORDS: Wavelet compression, quality evalua-tion, Moran test
T
HE PICTURE ARCHIVING ANDCOMMUNICATION SYSTEM1 (PACS)
utilizes digital technologies for the acquisition, storage, and transmission of radiological im-ages. One of the major difficulties in operating a digital radiology facility is the sheer volume of image data that must be handled. Although recent advances in computer technology have increased the data storage capacity and com-munication speeds, they alone are not sufficient to overcome this problem. Image compression techniques can be employed to reduce the data volume into a more manageable size without significantly compromising image quality.
Dig-ital image compression2 is divided into two categories defined as lossless and lossy. Lossless techniques3include run lengthencoding, Huff-man coding, differential pulse code modulation, and Lempel-Ziv, which enable the complete image to be reconstructed from the compressed data set as a perfect reproduction of the origi-nal. Lossless compression can reduce the image size only by a factor of 2 to 3. Muchhigher compression ratios are desirable to produce a more substantial compression impact. Lossy techniques enable significantly higher compres-sion levels withslight quality degradation.
Quality evaluation can be performed subjec-tively or objecsubjec-tively. The receiver operating characteristic (ROC) analysis is the dominant technique for subjectively evaluating image quality. In an ROC study,4 radiologists are asked to review compressed images withor without an abnormality and to provide a binary decision along withtheir degree of certainty. The diagnostic accuracies of these images were then compared with the original images. The ROC analyses are expensive and
time-consum-1
From the Department of Nuclear Sciences, National Tsing-Hua University, Hsinchu 30043, Taiwan; 2Health Physics Division, Institute of Nuclear Energy Research, Atomic Energy Council, Taiwan;3School of Medical Tech-nology, Kaohsiung Medical University, Taiwan; and4Physics Division, Institute of Nuclear Energy Research, Atomic En-ergy Council, Taiwan.
Correspondence to: Keh-Shih Chuang, Ph.D., Department of Nuclear Sciences, National Tsing-Hua University, Hsinchu 30043, Taiwan, tel: 886-3-574-2681; fax: 886-3-571-8649; e-mail [email protected]
Copyright 2003 by SCAR (Society for Computer Ap-plications in Radiology)
Online publication 2 October 2003 doi: 10.1007/s10278-003-1652-0
squared error (NMSE) is the most commonly used method to measure the quality changes in compressed images.6-11The NMSE is a measure of the image difference that is formed by taking the mean of the squared differences between all corresponding pixels in the original and the compressed images. The guidance document for PACS requires that manufacturers report to PDA the NMSE of their lossy compression technique.12 The NMSE is sensitive to degra-dation, withalterations in its value depending on the image content and degree of degrada-tion. However, NMSE does not provide any information regarding the type of loss that causes the quality deterioration, and it does not correlate well withsubjective quality measure-ments.
The purpose of this study was to establish a quality degradation model in lossy compression and present a method for measuring the image quality. Based on this model, the optimal compression can be determined for eachimage before compression is performed. In the fol-lowing paragraphs we briefly describe the irre-versible compression method. The method for using the Moran I test to measure the quality change is then introduced. The measurement results from various medical images are then be presented.
IRREVERSIBLE IMAGE COMPRESSION Irreversible image compression2 involves three processes: (i) image transformation, (ii) quantization, and (iii) entropy encoding. Image transformation, also referred to as decorrela-tion, performs transformation (mostly cosine
data volume.
Quantization is irreversible, and it is the cause of quality loss in compressed images. As the compression ratio increases, a greater number of high-frequency components are re-moved and the image becomes smoother. The major contributing factors to the high-fre-quency parts are statistical noise and structural edges. The quality loss is caused by the effects of denoising and edge blurring, and it can be quantified by measuring the smoothness of the compressed image.
MORAN I TEST
The Moran I test15,16 has been employed to evaluate the spatial autocorrelation of mapped data. In this study, it was used to measure the smoothness in an image. The Moran coefficient Ifor a pixel is calculated as:
I¼ P rc j¼1 P rc i¼1 dijðfi ffÞðfj ffÞ=S0 P rc i¼1 ðfi ffÞ2=N ; ð1Þ
where fiis the gray level of pixel i, ffis the mean gray level inside a r · c window centered on that pixel, dij= 1 if pixels i and j are adjacent, and 0 otherwise. Further, S0= 4rc) 2r ) 2c, is the number of contiguous pairs inside the window, and N (= r· c) is the total number of pixels. If the pixels inside the window are ran-domly distributed, the variable I can be ap-proximated using a normal distribution (when N is large enough) with mean and variance given by
m¼ 1=ðN 1Þ ð2Þ and
where K¼ NPð fi ffÞ4=½Pðfi ffÞ22, S1= 2S0, and S2= 8(8rc) 7r ) 7c + 4). For a smoothre-gion, the gray levels of adjacent pixels are more or less the same and the calculated I is larger. Note that I = 1 when all pixels have the same gray levels. The standardized normal statistic
z¼I m
r ð4Þ
is the smoothness measurement around a pixel. Thus, the Moran I test can be used to detect the quality change in a compressed image caused by the smoothing effects.
Z-HISTOGRAM AND PEAK RATIO The variation around a pixel is measured by calculating the Moran z value for a 9 · 9 win-dow centered on that pixel. The winwin-dow size is selected such that the result is statistically meaningful, and yet it is small enoughto reflect local variation. The histogram of the z values for all pixels in the image can be used to rep-resent the quality of the image. Figure 1 shows the z-histogram of a typical magnetic resonance
imaging (MRI) head image. There are two peaks in the histogram. The lower z peak cor-responds to the areas outside the skull that are predominantly noise. The high z peaks at about z= 9 (equivalent to I = 0.8) representing most pixels of the image are uniformly distrib-uted structures. In general, the quality evalua-tion of an image is performed on regions of interest and, in this case, inside the skull. Thus, we use only pixels inside the skull for head image studies. Figure 2 shows the z-histogram of the same MRI image for various degrees of compression. As the compression ratio in-creases, the image becomes smoother and the histogram curve shifts toward higher z regions and its peak value increases correspondingly. The peak ratio is defined as the ratio of peak values of the z-histogram between the com-pressed and original images. The peak ratio increases withcompression ratio and can serve as an indication of image degradation in lossy compression. The peak ratio is selected over other parameters like histogram mean ( ZZ) for its higher sensitivity to the change of compres-sion ratio.
Fig 1. The z-histogram of a typical MRI head image. Fig 2. The z-histogram of the same MRI image shown in Figure 1 (for pixels inside the skull only) at various degrees of compression. The peak ratio is defined as the ratio of peak height between the compressed and original images.
r2¼N½ðN 2 3N þ 3ÞS 1 NS2þ 3S20 K½NðN 1ÞS1 2NS2þ 6S20 ðN 1ÞðN 2ÞðN 3ÞS2 0 m2; ð3Þ
IMAGE DATA
Numerous images for eachdigital modality (body CT, head MRI T1 weighted (T1W), head MRI T2W, and chest CR) were randomly se-lected from different patients for this study. All images were compressed using wavelet compres-sion software ‘‘Apollo’’ (Pegasus Imaging Cor-poration, Tampa, FL). The Pegasus wavelet encoding is similar to that used in the JPEG 2000 standard. All measurements were performed for images withcompression ratios of 59, 49, 35, 30, 25, 23, 20, 18, 16, 14, 12, 10, 8, 7, and 5.
RESULTS
A CT image was used first to test the edge blurring and denoising effect on the peak ratio.
The image was filtered using an average filter to simulate the edge blurring effects. The filter window sizes can be adjusted to produce images withvarious degrees of edge blurring. The peak ratios of the z histogram and NMSE between eachfiltered images and original image were calculated. Figure 3 plots the peak ratio and NMSE as a function of the window size. The peak ratio was demonstrated to increase with the increasing degree of edge blurring. This is highly consistent with NMSE.
Because noise occupies the least significant bit planes, we can simulate the denoising effect by nullifying those bits. When a greater number
Fig 3. The peak ratio (s) and NMSE (·) versus window size of the average filter to simulate the effect of edge blurring.
Fig 4. The peak ratio (s) and NMSE (·) versus number of bits removed to simulate the effect of denoising.
Fig 5. The peak ratio of the body CT images as a function of compression ratio.
Fig 6. The peak ratio of the head MRI T1W images (for pixels inside the skull only) as a function of compression ratio.
of least significant bit planes is zeroed, more noise is smoothed out. We gradually increase the number of bit planes to be nullified and perform the Moran test on these denoised im-ages. Figure 4 shows the peak ratios as a function of number of bit planes nullified. The peak ratio decreases as the number of bits in-creases; ie, the peak ratio decreases with the degree of denoising. Note that the change in peak ratio in denoising the bit planes is much smaller than the blurring effects from the av-eraging filter. The NMSEs of the denoising images are also plotted in Figure 4. The NMSE increases withthe number of bit planes nulli-fied. The NMSE only calculates the sum of er-rors between corresponding pixels. It does not provide any information regarding the type of loss that causes the quality deterioration. The Moran peak ratio shows different trends for the effects of smoothing or denoising.
Figures 5-8 show the peak ratio for the CT, MRT1, MRT2, and CR images, respectively, as a function of the compression ratio. Note that at a low compression ratio, all of the peak ratio curves are nearly constant and they actually appear hollow for all images except CT. It has been reported17,18 that radiologists prefer im-ages processed withlow levels of compression. This preference can be attributed to the ‘‘de-noising’’ effect of the compression algorithm at low levels. At high compression ratios, the im-age qualities deteriorate mainly because of the blurring and the curves ascend linearly with the compression ratio.
DISCUSSION AND CONCLUSIONS For lossy compression in medical imaging, it is natural to question whether any clinically important information has been compromised. To solve this, a reliable way to quantify the quality degradation in compression is neces-sary. The image compression affects image quality through edge blurring and denoising. Both effects cause changes in the image smoothness. The Moran test is a measurement of the spatial correlation and is a good indica-tion of quality smoothness. We used the peak Moran test ratio to measure the quality changes resulting from compression. The results were in good agreement withthe ROC study. It is concluded that the Moran I test is a powerful tool for measuring quality degradation in image compression.
Note that the shapes of the peak ratio curves from different modalities are similar. This means that the compression effects on images are the same for all modalities. However, the degree of blurring caused by compression is dependent on the contrast of the original image; the blurring effects are less obvious for images with higher contrast. The CR images are formed using x-ray projection. X-ray contrasts are smaller than tomographic images. The CT body contrast is not as good as that for head MRI. As a result, the CR images have the highest slopes, followed by CT and MRI.
For CR chest images, an increasing number of reports have suggested that compression
Fig 7. The peak ratio of the head MRI T2W images (for pixels inside the skull only) as a function of compression ratio.
Fig 8. The peak ratio of the chest CR images as a function of compression ratio.
ity is so coherent that the peak ratio curves go downward and then upward in response to the compression ratio, indicating the existence of optimal compression. The beginning point of the peak ratio curve is a good approximation of the inherent noise for each image modality. By drawing a horizontal line from that point, the optimal compression ratio can be estimated from the intersection of the curve with the line. From this, we can reason that the optimal compression ratios are around 15 for most modalities. This result is close to the previous results of a study.2
ACKNOWLEDGMENTS
This work is supported in part by research grant NSC89-2314-B007 from the National Science Council, Taiwan. The image data were provided by S. C. Kuo and Alex Hsu of the Department of Radiation Oncology, Chung-Shan Medical & Dental College Hospital, Taiwan.
REFERENCES
1. Huang HK: PACS—Picture Archiving and Commu-nication Systems in Biomedical Imaging. New York, VCH Publishers, 1996
2. Wong S, Zaremba L, Gooden D, et al: Radiologic image compression—a review. Proc IEEE 83:194-219, 1995 3. Chen ZD, Chang RF, Kuo WJ: Adaptive predictive multiplicative autoregressive model for medical image compression. IEEE Trans Med Imaging 18:181-184, 1999
4. Swets JA: ROC analysis applied to the evaluation of medical imaging techniques. Invest Radiol 14:109-121, 1979 5. Ji TL, Sundareshan MK, Roehrig H: Adaptive image contrast enhancement based on human visual properties. IEEE Trans Med Imaging 13:573-586, 1994
diol 73:1306-1312, 2000
11. Good WF, Gur D, Feist JH, et al: Subjective and objective assessment of image quality—a comparison. J Digit Imaging 7:77-78, 1994
12. Center for Devices and Radiological Health, the US Food and Drug Agency. Guidance for the submission of premarket notification for medical image management de-vices. Washington DC, 2000
13. DeVore R, JawerthB, Lucier B: Image compression through wavelet transform coding. IEEE Trans Info Theory 38:719-746, 1992
14. Goldberg MA, Pivovarov M, Mayo-SmithWW, et al: Application of wavelet compression to digitized radio-graphs. AJR Am J Roentgenol 163:463-468, 1994
15. Cliff AD, Ord JK: Spatial Process: Methods and Applications. London: Pion, 1981
16. Chuang KS, Liu BJ, Huang HK, et al: Noise content analysis in clinical digital images. Radiographics 14:397-403, 1994
17. SmithI, Roszkowski A, Slaughter R, et al: Accept-able levels of digital image compression in chest radiology. Australas Radiol 44:32-35, 2000
18. Savcenko V, Erickson BJ, Palisson PM, et al: De-tection of subtle abnormalities on chest radiographs after irreversible compression. Radiology 206:609-616, 1998
19. MacMahon H, et al: Data compression: effect on diagnostic accuracy in digital chest radiography. Radiology 178:175-179, 1991
20. Zhao B, Schwarz LH, Kijewski PK: Effects of lossy compression on lesion detection: predictions of the non-prewhitening matched filter. Med Phys 25:1621-1624, 1998
21. Kido S, Ikezoe J, KondohH, et al: Detection of subtle interstitial abnormalities of the lungs on digital chest radiographs: acceptable data compression ratio. AJR Am J Roentgenol 167:111-115, 1996
22. Slone RM, Foos DH, et al: Assessment of visually lossless irreversible image compression: comparison of three methods by using an image-comparison workstation. Ra-diology 215:543-553, 2000
23. Cox GG, Cook LT, Insana MP, et al: The effects of lossy compression on the detection of subtle pulmonary nodules. Med Phys 23:127-132, 1996
