Predicting the fidelity of JPEG2000 compressed CT images using DICOM header information.

PURPOSE To propose multiple logistic regression (MLR) and artificial neural network (ANN) models constructed using digital imaging and communications in medicine (DICOM) header information in predicting the fidelity of Joint Photographic Experts Group (JPEG) 2000 compressed abdomen computed tomography (CT) images. METHODS Our institutional review board approved this study and waived informed patient consent. Using a JPEG2000 algorithm, 360 abdomen CT images were compressed reversibly (n = 48, as negative control) or irreversibly (n = 312) to one of different compression ratios (CRs) ranging from 4:1 to 10:1. Five radiologists independently determined whether the original and compressed images were distinguishable or indistinguishable. The 312 irreversibly compressed images were divided randomly into training (n = 156) and testing (n = 156) sets. The MLR and ANN models were constructed regarding the DICOM header information as independent variables and the pooled radiologists' responses as dependent variable. As independent variables, we selected the CR (DICOM tag number: 0028, 2112), effective tube current-time product (0018, 9332), section thickness (0018, 0050), and field of view (0018, 0090) among the DICOM tags. Using the training set, an optimal subset of independent variables was determined by backward stepwise selection in a four-fold cross-validation scheme. The MLR and ANN models were constructed with the determined independent variables using the training set. The models were then evaluated on the testing set by using receiver-operating-characteristic (ROC) analysis regarding the radiologists' pooled responses as the reference standard and by measuring Spearman rank correlation between the model prediction and the number of radiologists who rated the two images as distinguishable. RESULTS The CR and section thickness were determined as the optimal independent variables. The areas under the ROC curve for the MLR and ANN predictions were 0.91 (95% CI; 0.86, 0.95) and 0.92 (0.87, 0.96), respectively. The correlation coefficients of the MLR and ANN predictions with the number of radiologists who responded as distinguishable were 0.76 (0.69, 0.82, p < 0.001) and 0.78 (0.71, 0.83, p < 0.001), respectively. CONCLUSIONS The MLR and ANN models constructed using the DICOM header information offer promise in predicting the fidelity of JPEG2000 compressed abdomen CT images.

[1]  S. Kumano,et al.  Multidetector CT: diagnostic impact of slice thickness on detection of hypervascular hepatocellular carcinoma. , 2002, AJR. American journal of roentgenology.

[2]  Colin Studholme,et al.  Intersection Based Motion Correction of Multislice MRI for 3-D in Utero Fetal Brain Image Formation , 2010, IEEE Transactions on Medical Imaging.

[3]  M. Essig,et al.  Clinical experience with multihance in CNS imaging , 2003, European Radiology.

[4]  Bo Hyoung Kim,et al.  Irreversible JPEG 2000 compression of abdominal CT for primary interpretation: assessment of visually lossless threshold , 2006, European Radiology.

[5]  Thomas Richter,et al.  JPEG2000 3D compression vs. 2D compression: an assessment of artifact amount and computing time in compressing thin-section abdomen CT images. , 2009, Medical physics.

[6]  Jin Mo Goo,et al.  Radiation dose modulation techniques in the multidetector CT era: from basics to practice. , 2008, Radiographics : a review publication of the Radiological Society of North America, Inc.

[7]  David R Pickens,et al.  Extracting data from a DICOM file. , 2005, Medical physics.

[8]  Bohyoung Kim,et al.  Differences in compression artifacts on thin- and thick-section lung CT images. , 2008, AJR. American journal of roentgenology.

[9]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[10]  Heung-Sik Kang,et al.  Regional difference in compression artifacts in low-dose chest CT images: effects of mathematical and perceptual factors. , 2008, AJR. American journal of roentgenology.

[11]  A. Dupuy,et al.  Critical review of published microarray studies for cancer outcome and guidelines on statistical analysis and reporting. , 2007, Journal of the National Cancer Institute.

[12]  Rafal Mantiuk,et al.  Prediction of perceptible artifacts in JPEG 2000-compressed chest CT images using mathematical and perceptual quality metrics. , 2008, AJR. American journal of roentgenology.

[13]  Yi Wang,et al.  3-T navigator parallel-imaging coronary MR angiography: targeted-volume versus whole-heart acquisition. , 2008, AJR. American journal of roentgenology.

[14]  Bohyoung Kim,et al.  Regional variance of visually lossless threshold in compressed chest CT images: lung versus mediastinum and chest wall. , 2009, European journal of radiology.

[15]  Yeong-Gil Shin,et al.  Advantage in image fidelity and additional computing time of JPEG2000 3D in comparison to JPEG2000 in compressing abdomen CT image datasets of different section thicknesses. , 2010, Medical physics.

[16]  Bohyoung Kim,et al.  Definition of compression ratio: difference between two commercial JPEG2000 program libraries. , 2008, Telemedicine journal and e-health : the official journal of the American Telemedicine Association.

[17]  Lucila Ohno-Machado,et al.  Logistic regression and artificial neural network classification models: a methodology review , 2002, J. Biomed. Informatics.

[18]  Heung-Sik Kang,et al.  Objective index of image fidelity for JPEG2000 compressed body CT images. , 2009, Medical physics.

[19]  Dianna D Cody,et al.  AAPM/RSNA physics tutorial for residents: topics in CT. Image processing in CT. , 2002, Radiographics : a review publication of the Radiological Society of North America, Inc.

[20]  Yeong-Gil Shin,et al.  Comparison of three image comparison methods for the visual assessment of the image fidelity of compressed computed tomography images. , 2011, Medical physics.

[21]  Bohyoung Kim,et al.  JPEG 2000 compression of abdominal CT: difference in tolerance between thin- and thick-section images. , 2007, AJR. American journal of roentgenology.

[22]  Michael F McNitt-Gray,et al.  AAPM/RSNA Physics Tutorial for Residents: Topics in CT. Radiation dose in CT. , 2002, Radiographics : a review publication of the Radiological Society of North America, Inc.

[23]  Rafal Mantiuk,et al.  Prediction of perceptible artifacts in JPEG2000 compressed abdomen CT images using a perceptual image quality metric. , 2008, Academic radiology.

[24]  B. Choi,et al.  Appropriateness of a donor liver with respect to macrosteatosis: application of artificial neural networks to US images--initial experience. , 2005, Radiology.

[25]  Edward Muka,et al.  Irreversible JPEG compression of digital chest radiographs for primary interpretation: assessment of visually lossless threshold. , 2003, Radiology.

[26]  J. Baker,et al.  Breast mass lesions: computer-aided diagnosis models with mammographic and sonographic descriptors. , 2007, Radiology.

[27]  Helen Hong,et al.  Managing the CT Data Explosion: Initial Experiences of Archiving Volumetric Datasets in a Mini-PACS , 2005, Journal of Digital Imaging.

[28]  J V Tu,et al.  Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. , 1996, Journal of clinical epidemiology.

[29]  G. Rubin,et al.  Data explosion: the challenge of multidetector-row CT. , 2000, European journal of radiology.

[30]  Seokyung Hahn,et al.  Computed Tomography Diagnosis of Acute Appendicitis: Advantages of Reviewing Thin-section Datasets using Sliding Slab Average Intensity Projection Technique , 2006, Investigative radiology.

[31]  Helen Hong,et al.  Summation or Axial Slab Average Intensity Projection of Abdominal Thin-section CT Datasets: Can They Substitute for the Primary Reconstruction from Raw Projection Data? , 2007, Journal of Digital Imaging.

[32]  J. Hanley,et al.  A method of comparing the areas under receiver operating characteristic curves derived from the same cases. , 1983, Radiology.

[33]  Oguzhan Alagoz,et al.  Informatics in radiology: comparison of logistic regression and artificial neural network models in breast cancer risk estimation. , 2010, Radiographics : a review publication of the Radiological Society of North America, Inc.

[34]  Patrick van der Smagt,et al.  Introduction to neural networks , 1995, The Lancet.

[35]  Francis R Verdun,et al.  Detection of low-contrast objects: experimental comparison of single- and multi-detector row CT with a phantom. , 2002, Radiology.

[36]  Thomas Richter,et al.  A Comparison of Three Image Fidelity Metrics of Different Computational Principles for JPEG2000 Compressed Abdomen CT Images , 2010, IEEE Transactions on Medical Imaging.

[37]  Rafal Mantiuk,et al.  Artifacts in slab average-intensity-projection images reformatted from JPEG 2000 compressed thin-section abdominal CT data sets. , 2008, AJR. American journal of roentgenology.

[38]  April Khademi,et al.  Pan-Canadian Evaluation of Irreversible Compression Ratios (“Lossy” Compression) for Development of National Guidelines , 2008, Journal of Digital Imaging.