Feature clustering of intracranial pressure time series for alarm function estimation in traumatic brain injury.

OBJECTIVE: The conventional application of intracranial pressure (ICP) monitoring of traumatic brain injury (TBI) patients consists merely in the acquisition of ICP values in discrete time and their comparison to the established ICP threshold. An exceeding of this threshold triggers a special emergency treatment protocol. This paper addresses the possibility of making use of the rich information latent in the ICP records of known vital and fatal outcomes gathered during real clinical practice of treating TBI patients. Our assumption was that the proposed algorithmic procedure derived from this information could, in addition to ICP monitoring itself, provide a complementary added value. This might help clinicians to make better decisions during a patient's treatment. APPROACH: We concentrated on studying specific clustering schemes for subsequences of ICP time series. The clusterization problem was formulated for feature vectors which are introduced to represent ICP time subsequences. The ICP transformation to a feature space uses global and local definitions of time subsequences. For clusterization itself, we adopted hierarchical Gaussian mixture models (hGMMs). By using posterior probabilities of the clusters, we introduced three novel alarm functions. We explored two alternative methods of searching for optimum alarm function thresholds (ROC analysis and a novel efficiency measure). MAIN RESULTS: We performed extensive cross-validation experiments on a clinical retrospective data set. The results of the optimization over several hGMMs, various feature space dimensionality and all the types of the novel alarm functions show the potential of the novel alarm functions for supplementing conventional ICP monitoring. SIGNIFICANCE: In conclusion, the paper provides a prospective extended ICP monitoring technique for real TBI patients, based on the proposed methodology of ICP subsequence clustering and thresholding of the optimum novel alarm function.

Improvement of band segmentation in Epo images via column shift transformation with cost functions.

In recent years, the development of methodology and laboratory techniques for doping control (DC) of recombinant erythropoietin (rEpo) has become one of the most important topics pursued by doping control laboratories accredited by World Anti-Doping Agency (WADA). The software system GASepo has been developed within the international WADA project as a support for Epo doping control. Although a great number of functions for automatic image processing have been involved in this software, for Epo images with considerably distorted bands additional effort is required from the user to interactively correct the results of improper band segmentation. In this paper a problem of geometrically distorted bands is addressed from the viewpoint of how to transform the lanes in distorted Epo images in order to reach better band segmentation. A method of band straightening via column shift transformation has been proposed that is formulated as an optimization procedure with cost functions. The method involves several novel approaches: two-stage optimization procedure, four cost functions and selection of relevant columns. The developed band straightening algorithm (BSA) has been tested on real Epo images with distorted bands. Based on the evaluation scheme involving the GASepo software itself a recommendation is made for implementation of the method with the cost function based on correlation matrix. Estimates of computational complexity of the individual steps of BSA are also given.

GASepo-a software solution for quantitative analysis of digital images in Epo doping control.

A software has been developed that is aimed at quantitative analysis of images acquired by isoelectric focusing and double blotting procedures used for recombinant erythropoietin doping control. It represents a unified and easy-to-use tool for Epo doping experts in WADA accredited laboratories. It is based on image segmentation philosophy that enables identification of individual bands whose characteristics are needed for evaluation of the Epo doping positivity criteria. Several modules implemented in the GASepo software include an original know how, in particular, the method of robust calculation of the cut-off line, band segmentation and classification algorithms. GASepo is being used in several doping control laboratories worldwide.

Quantitative analysis of images in erythropoietin doping control.

The software system GASepo has been developed as a tool to visualise and analyse doping (searching for performance-enhancing drugs in sports) with recombinant erythropoietin (rEpo). Digital images derived from the separation of Epo isoforms in gels by iso-electric focusing followed by double blotting and chemiluminescence detection contain spots (bands) of characteristic shape and positions. For Epo doping control, these have to be analysed and evaluated. A relevant element of the analysis is calculation of the reference cutoff-line (COL) to which the doping positivity criterion is related. Based on analysis of the previous method used, a novel method for the COL calculation was developed and validated. The methodology was based on generation of a partition of the image (lane) being processed into a system of adjacent subimages (blocks) and quantitative characterisation of intensity variability within these blocks by a suitably defined measure. The coordinate of the proper COL position, separating the image part with recombinant Epo bands from the homogeneous background, was calculated by minimisation of the difference function of the measures for two adjacent image blocks. The proposed method was tested on real Epo images and validated on a synthetic phantom of the Epo image. The deviations of the COL position are limited by a 7 pixel-wide corridor. For the cases of noise with standard deviations 2300 and 2600, the deviations did not exceed 2 pixels over the whole range of the lane width values. Testing the method on 50 real Epo images originating from different doping-control laboratories worldwide showed its robust behaviour in Epo images.

An alternative method for electrophoretic gel image analysis in the GelMaster software.

A novel methodology of electrophoretic gel image analysis has been proposed that is based on two-dimensional image processing methods instead of previously used one-dimensional Gaussian deconvolution. The crucial problem of the analysis of imperfect gels, that consists in band detection, is solved using the algorithms of band boundary detection and intensity homogeneity indication. The template approach represents the core element of the developed algorithms. The GelMaster software system has been developed in which the novel algorithms are implemented. It involves two-stage interaction with the user: detection of the true bands and deleting the false band detections. The main features of the GelMaster system and the most important algorithms are described.

Adaptive smoothing of MR brain images by 3D geometry-driven diffusion.

The paper deals with the iterative three-dimensional (3D) smoothing of tomograms acquired by fast Magnetic Resonance (MR) imaging methods. The smoothing method explored, which is aimed basically at the improvement of 3D visualization quality, uses the physical concept of geometry-driven diffusion with a variable conductance function, based on a specific measure of the 3D neighborhood homogeneity. A novel stopping criterion is proposed for iterative 3D diffusion processing. A study of the transition from 2D to 3D algorithms is carried out. The main structure of the program implementation of the smoothing algorithms developed is described. Three smoothing/filtering methods, aimed at the improvement of 3D visualization of MR tomograms of the brain, are quantitatively and visually compared using real 3D MR images. The results of computer simulations with 3D smoothing, segmentation and visualization are presented and discussed.

Improvement of 3D visualization of the brain using anisotropic diffusion smoothing of MR data.

This paper deals with an iterative 3D smoothing of MR images acquired by fast 3D Flash sequences. The method explored, which is aimed at improving 3D visualization quality, uses the physical concept of anisotropic diffusion. A novel iteration stopping criterion is proposed and tested. The results of computer experiments with 3D smoothing, segmentation, and visualization using actual MR brain data are presented and discussed.

[Generation and processing of digital images in radiodiagnosis]. / Tvorba a spracovanie digitálnych obrazov v rádiodiagnostike.

The paper describes universal principles of diagnostic imaging. The attention is focused particularly on digital image generation in medicine. The methodology of display visualization of measured data is discussed. The problems of spatial relation representation and visual perception of image brightness are mentioned. The methodological issues of digital image processing (DIP) are discussed, particularly the relation of DIP to the other related disciplines, fundamental tasks in DIP and classification of DIP operations from the computational viewpoint. The following examples of applying DIP operations in diagnostic radiology are overviewed: local contrast enhancement in digital image, spatial filtering, quantitative texture analysis, synthesis of the 3D pseudospatial image based on the 2D tomogram set, multimodal processing of medical images. New trends of application of DIP methods in diagnostic radiology are outlined: evaluation of the diagnostic efficiency of DIP operations by means of ROC analysis, construction of knowledge-based systems of DIP in medicine. (Fig. 12, Ref. 26.)

On the possibility of direct Fourier reconstruction from divergent-beam projections.

A great number of tomographic systems, especially those equipped with fast data acquisition techniques, scan the objects investigated by divergent (fan) X-ray beams. Fan-beam projection data require special reconstruction techniques to be implemented. Among reconstruction techniques from parallel projection data, the direct Fourier method (DFM) proved to be one of the most promising ones, especially for high-speed image reconstruction in the high-end 3-D and dynamic tomographic systems. The goal of this work is to answer the topical question: how would direct use of the DFM influence the quality of image reconstruction from the fan-beam data? The formula describing the error caused by such an approximation is derived. The conclusions deduced from the formula are confirmed by computer simulations. The boundary values of data acquisition geometry parameters have been estimated for the case of using the DFM without recalculating the fan-beam data.

A high-speed reconstruction from projections using direct Fourier method with optimized parameters-an experimental analysis.

Some issues of the direct Fourier method (DFM) implementation are discussed. A hybrid spline-linear interpolation for the DFM is proposed. The results of comprehensive simulation research are presented. The following reconstruction problems and parameters are emphasized: interpolation, increasing the radial density of the polar raster, filtering, the 2-D inverse Fourier transformation dimension, and considering the cases of noiseless and noisy input data. For the a priori prescribed resolution of the reconstructed image, values of reconstruction parameters have been determined which are optimal with regard to reconstruction quality and computation cost. The computational requirements of the DFM algorithm which correspond to distinct interpolation schemes are compared to one another for CT and MR tomography, respectively. The estimations obtained are compared to computational characteristics of the convolution backprojection method.