MUCKE aims to mine a large volume of images, to structure them conceptually and to use this conceptual structuring in order to improve large-scale image retrieval. The last decade witnessed important progress concerning low-level image representations. However, there are a number problems which need to be solved in order to unleash the full potential of image mining in applications. The central problem with low-level representations is the mismatch between them and the human interpretation of image content. This problem can be instantiated, for instance, by the incapability of existing descriptors to capture spatial relationships between the concepts represented or by their incapability to convey an explanation of why two images are similar in a content-based image retrieval framework. We start by assessing existing local descriptors for image classification and by proposing to use co-occurrence matrices to better capture spatial relationships in images. The main focus in MUCKE is on cleaning large scale Web image corpora and on proposing image representations which are closer to the human interpretation of images. Consequently, we introduce methods which tackle these two problems and compare results to state of the art methods. Note: some aspects of this deliverable are withheld at this time as they are pending review. Please contact the authors for a preview.

http://ieeexplore.ieee.org/iel2/4524/12820/00592460.pdf

Image Processing

Statistical Methods in Medical Research.

The traditional chest radiograph is still ubiquitous in clinical practice, and will likely remain so for quite some time. Yet, its interpretation is notoriously difficult. This explains the continued interest in computer-aided diagnosis for chest radiography. The purpose of this survey is to categorize and briefly review the literature on computer analysis of chest images, which comprises over 150 papers published in the last 30 years. Remaining challenges are indicated and some directions for future research are given.

/pdf/computer-aided-diagnosis-in-chest-radiography-a-survey-2g53b77kmg.pdf

Computer-aided diagnosis in chest radiography: a survey

Storage phosphor radiography is a digital technique that uses photo-stimulable phosphor screens to substitute for conventional screen-film combinations. While the technique is more than 15 years old, it is only recently that technological and economic aspects of these systems have become favourable enough to envisage a more widespread clinical application.

Storage phosphor radiography

Clinical experiences with computed radiography

PURPOSE To examine the extent to which digital luminescence radiography (DLR) can be used for the imaging of pulmonary nodules and interstitial lung disease in chest radiography without any loss of image quality. Additionally: to examine whether post-processing of image data can optimise the recognizability of varied image details. MATERIALS AND METHODS Detail perceptibility studies were performed on an anthropomorphic thorax phantom with simulated nodules and small linear and reticular details. Under standard conditions, digital luminescence radiographs were obtained in 7 different image modes, and these were compared with a 200-speed screen-film system. The detection of these systems was evaluated in an ROC analysis on the basis of 19,200 individual observations. RESULTS Edge enhancement or application of high-frequency-enhancing small filter kernels (S 5) slightly improves the detection of linear structures; however, the illustration of nodular details is markedly reduced. Larger filter kernels (S 20, S 40) make a definitive detection possible--not only of circular, but also of linear details. CONCLUSIONS Storage phosphor radiographs are equal to the tested analog screen-film-system. The optimization of post-processing can be helpful in the prevention of routine multiple documentations.

ROC-Analyse zur Bildnachverarbeitung digitaler Thoraxaufnahmen

Purpose: The aim of the study was to optimize the postprocessing of digital luminescence radiographs of interstitial infiltrates of the lung.Material and Methods: Ninety-seven patients with established diseases of the lung parenchyma were examined: 514 pathological details were investigated, using a 200-speed screen-film system and a storage phosphor plate. In digital postprocessing, the size of the filter kernel (S) varied between S 5 and S 70. Based upon 32896 individual evaluations, the image quality was evaluated in multifactorial variance analyses.Results: The digital standards were significantly inferior to the tested screen-film system. The large filter kernel S 70 and the small filter kernels S 5 and S 10, as well as an algorithm for edge enhancement, showed selective advantages in the depiction of nodular or small linear structures. For micronodular and linear structures, kernel sizes of S 20 and S 40 produced equally good results (p>0.05).Conclusion: Whenever optimized postprocessing is involved...

Effects of varying filter kernel sizes on the image quality of interstitial lung diseases.

The aim of this work was to examine the influence of the filter kernel size on the detectability of differing radiological findings in interstitial lung disease. In 97 patients with confirmed pulmonary fibrosis chest radiographs were obtained with a film-screen system of speed class 200 and with correspondingly exposed storage phosphorous plates. The size of the filter kernel used for the image postprocessing varied between sigma 5 and sigma 70. The detectability of interstitial lung changes was evaluated independently by eight readers on the basis of a defined rating system. The results were analysed using multifactorial analysis of variance with Scheffe test at a significance level of p = 0.05. Small kernel sizes (S 5, S 10) combined with high edge enhancement were only of benefit in the imaging of septal lines, but reduced the detectability of nodular and reticular structures. Good detail detectability of both micronodules and septal lines was obtained with a medium kernel size of sigma 40. Storage phosphor radiography utilizing the appropriate choice of postprocessing parameters provides equivalent image quality for evaluating interstitial lung changes compared with a modern film-screen technique.

Detection of radiographic findings in lung fibrosis using storage phosphor plates

Purpose: To compare the performances of 4 imaging systems in the diagnostic radiology of the chest: an asymmetric screen-film combination (a-SFC); a conventional screen-film combination of speed class 200 (SFC); a screen-film combination of speed class 200 with an aluminium lung filter; and digital luminescence radiographs (DLR).Material and Methods: Object detectability studies were carried out using an anthropomorphic chest phantom. the images were obtained under standard exposure conditions at 125 kVp. the detection performance of the various systems was evaluated by receiver operating characteristic (ROC) analysis on the basis of a total of 12 240 observations.Results: In the overall evaluation the a-SFC and the SFC plus lung filter received equally high ratings of 0.878±0.018 and 0.860±0.21, respectively (p>0.05). There was no significant difference between the 200-speed SFC and the storage phosphor radiographs. In the mediastinum all the tested systems were clearly superior to the 200-speed SFC (p<0...

Roc-analysis of detection performance by analogue and digital plain film systems in chest radiography

Digital luminescent radiography enables dual-energy subtraction imaging, because this computed system allows susbtraction of imaging data and image post processing, as for example special windowing or edge enhancement. In a special cassette a copper filter is placed between two imaging plates for energy separation by a single X-ray exposure. Image post-processing with subtraction of imaging data permits the elimination of either skeleton or soft tissue structures. The influence of filter thickness, tube voltage and the X-ray exposure dosage on image quality is examined by the use of an anthropomorphic quality is examined by the use of an anthropomorphic phantom of the chest. According to our initial results dual energy subtraction imaging in one-shot-technique seems to be useful in the diagnostics of skeletal lesions and especially pulmonary nodules.

The influence of tube voltage, X-ray exposure rate and thickness of filter material on image quality with dual-energy subtraction imaging in chest diagnostics

P. Gocke

Papers

ROC-Analyse zur Bildnachverarbeitung digitaler Thoraxaufnahmen

Effects of varying filter kernel sizes on the image quality of interstitial lung diseases.

Detection of radiographic findings in lung fibrosis using storage phosphor plates

Roc-analysis of detection performance by analogue and digital plain film systems in chest radiography

The influence of tube voltage, X-ray exposure rate and thickness of filter material on image quality with dual-energy subtraction imaging in chest diagnostics