This paper explains the concepts of knowledge management and how these apply to healthcare. It then presents a number of knowledge management case studies in order to examine how and where knowledge management initiatives might bring benefits to healthcare organisations. In order to present a more complete picture of knowledge management, the arguments of a number of critics of the knowledge management approach are presented and analysed.

This paper describes a hybrid segmentation method of cytological images for their analysis by means of Hough transform and Active Contours methods. One also can find there a short description of image preprocessing that was recommended for preliminary experiments organisation. The results were collected by a specifically prepared benchmarking database.

This paper is an extended version of the MIT"06 conference contribution. During the conference, many inquiries about the used techniques were performed. Hence, in the paper some parts of investigations were explained and discussed, with greater accuracy. It is shown that the computer applications can be controlled by a human voice. The computer controlling processes are available by means of utterance of isolated words, where application events with the aid of user"s voice can be serviced. The voice usage can be convenient for blind or partially sighted users or for persons with limb paresis. The Microsoft application events, by means of the practicable Microsoft Windows firmware MSAA® technology can be analysed. Such technology, together with isolated word descriptors, as voice recognition system, has been presented.

Applications of image registration in parametric imaging are investigated. The manual and automatic image registration methods have been used for image to image registration in sequences to correct movement artefacts in reconstructed parametric objects. Additionally the registration methods were used for multimodal visualisation of structural and parametric objects. The achieved results proved that the automatic image to image registration, for motion mechanisms correction, in the parametric model improves the quality of images. The multimodal visualisation of structural MRI images and parametric DSC-MRI images, enables to correlate the local dynamic changes with all morphological features.

Brain perfusion imaging using Dynamic Susceptibility Contrast Magnetic Resonance Imaging is very promising method since it can be easily implemented as a standard contrast-based MRI procedure. Quantitative brain perfusion description by DSC-MRI data post processing requires validation. Different validation analysis was performed to verify the influence of a bolus dispersion, delay, low SNR and calculation procedures on final perfusion parameter values. The results indicate that quantitative description of brain perfusion using DSC-MRI is possible and can be acceptable with accuracy about 10%.

The paper presents aspects connected with development of a computer system for registration and fusion of multimodal images for orthodontic purposes. The idea of the system, of the multimodal image registration, lies in 3D surfaces registering, reconstructed from the sets of MRI or CT planar slices. The system enables the user to perform registration using different approaches based on localised landmarks, based on selected regions of 3D surfaces or based on automatic matching of the whole surfaces. The rigid-body transformation found during the registration is then used to transform the set of slices to generate images that can be fused. The usage of the system was illustrated and the accuracy of the registration was assessed.

A fuzzy approach to segmentation of the cruciate ligaments of the knee joint and a three dimensional visualisation method are presented in this paper. The cruciate ligaments are the major stabilizers of the knee. The ligaments injuries are common nowadays and a correct diagnostics, preceding the surgical therapy is a very important task. Segmentation of the ligaments is difficult due to a poor visibility of edges in some cases of injuries and their appearance on a small number of slides at Magnetic Resonance Imaging (MRI). The method described here is based on fuzzy connectedness principles. It creates a fuzzy connectivity scene by assigning a strength of connectedness to each possible path between some predefined seed point and any other image element. Then such scene is thresholded to produce final segmentation result. The conventional fuzzy connectedness method with Dijkstra algorithm for creating the fuzzy connectivity scene has been implemented in a 3D space. The object, being the result of segmentation process, is visualised in the Visualisation Toolkit (VTK) environment. The method has been tested on a set of images. An example of its performance is shown along with some plans for future research.

In this paper the off-line type signature analysis have been presented. The signature recognition is composed of some features. Different influences of such features were tested and stated. Proposed approach gives good signature recognition level, hence described method can be used in many areas, for example in biometric authentication, as biometric computer protection or as method of the analysis of person"s behaviour changes.

An electrocardiogram (ECG) is the prime tool in non-invasive cardiac electrophysiology and has a prime function in the screening and diagnosis of cardiovascular diseases. However one of the greatest problems is that usually recording an electrical activity of the heart is performed in the presence of noise. The paper presents empirical Bayesian approach to problem of signal averaging which is commonly used to extract a useful signal distorted by a noise. The averaging is especially useful for biomedical signal such as ECG signal, where the spectra of the signal and noise significantly overlap. In reality the variability of noise can be observed, with power from cycle to cycle, which is motivation for weighted averaging methods usage. It is demonstrated that by exploiting a probabilistic Bayesian learning framework, it can be derived accurate prediction models offering significant additional advantage, namely automatic estimation of ‘nuisance’ parameters. Performance of the new method is experimentally compared to the traditional averaging by using arithmetic mean and weighted averaging method based on criterion function minimization.

This paper describes a new method for linear phase finite impulse response (FIR) filters design. This new approach, based on the ε-insensitive loss function, allows the design process to take into account not only constraints specified in the frequency domain, but also constraints on the output, time domain, signal. The performances of the proposed approach are shortly illustrated by a high-pass filter for ECG baseline wander reduction.

Robust filtering is a very promising area in application of biomedical signal processing. Signals are usually recorded with noise, which has various characteristics of baseline wander to very impulsive nature. The robust technique has been recently proposed as the tool to eliminate outliers in data samples. The main purpose of this paper is to present mean-median filters in application of ECG signal processing. The presented filter is evaluated in the presence of real muscle noise and simulated impulsive noise as a Gaussian-Laplace mixture. In order to suppress a noise with the best possible means, the special expression is proposed. The measure of distortions, which are introduced to a signal after operation of filtering, is estimated using the normalized mean square error. This factor is used to compare a quality of considered filters. Experimental results show improved performance according to the reference filters.

In this paper the Iris Finder program is presented for the first time. It is software for reliable iris localisation in images taken under visible light. The program is intended for researches on a system for automatic persons identification based on iris pattern. Performance of the software was verified on two databases. For the first database the program correctly localises iris for all 141 images. For the second database the iris was incorrectly localised only for 4 images from the entire database containing 1205 units.

The paper presents influence of mechanical surface damage on corrosion resistance of plates made of Cr-Ni-Mo stainless steel, used in funnel chest treatment. The surface of the steel was electrochemically polished and fitted. The surface damage is induced in the given deformation regions and is a potential reason of corrosion. The corrosion tests were realised by recording of anodic polarization curves with the use of the potentio-dynamic method. The VoltaLab® PGP 201 system for electrochemical tests was applied. Additionally, the tests showed that the structure of the steel the plates were made of, met the PN-ISO 5832-1 standard. On the basis of the obtained results it can be stated that that stainless steel can be applied in funnel chest treatment.

A new hybrid clustering method based on a fuzzy myriad is presented. The proposed method could be considered as a generalisation of the well known fuzzy c-means method (FCM) proposed by Bezdek. Existing modifications of the FCM method, such as conditional clustering or partial supervised clustering can be applied to determine the objective function of the proposed method.

The research on the human reading technique and text perception strategies needs objective evaluation methods. Our proposal consists in the use of scan-path recording and eye track processing results as estimators of reading skills and capability of error compensation. Visual tasks, being principal investigation tool, were completed by arrangement of texts for presentation, eye trajectory acquisition and assessment of comprehension degree. Our results show that gaze point statistics represent well the observer performance and skills in fast reading even in free-spotting-based visual tasks. Additionally, we reveal very high human tolerance for errors, outperforming any known optical character recognition software.

The paper concerns the implementation of some diagnostic measures describing gait disturbances produced by neurological diseases. The discussed contribution explains various experiments provided by the authors" works on Parotec System for Windows development [1]. The subject of the investigations was defined by many experiments carried-out at clinics, with gait"s characteristic features analysis, involving the fuzzy logic paradigms. The linguistic notations of diagnostic classes are computed dynamically in accordance with disease statistics, provided by the characteristics of the patient under investigation. The comprehensive system was recognised as a friendly user package, which insulates the user from the complex analysis of measures describing the physiological gait characteristic features. What is more, the application has the ability to be tuned more precisely after the data record size has grown. The software-units implementation provided the user with a selftuned system that enriches the application knowledge, during the data acquisition, while the set of the data records is still growing.

This study was based on observations of 117 patients suffering from motor disturbances. Among them 42 cases with hemiparetic syndrome, majority of them after cerebral stroke, 52 cases affected by acute sciatic neuralgia and 23 patients with recognition of Parkinson - disease symptoms. To the control group 16 healthy adults were selected, from medical staff of clinics. All subjects were examined using pedobarographic equipment - Parotec System for Windows (PSW) [1]. Based on these observations several pattern solutions have been introduced. They concern gait disturbances descriptions in three distinguished neurological diseases. These findings extracted a new data from the PSW records and options with new diagnostic techniques, based on the gait characteristics observation.

Diagnosing of morbid conditions by means of automatic tools supported by computers is a significant and often used element in modern medicine. Some examples of these tools are automatic conclusion-making units of Parotec System for Windows (PSW). In the initial period of PSW system implementation, the units were used for recognition of orthopaedic diseases on the basis of the patient"s walk and posture [15,17]. Subsequently, many additional options have been implemented, which have been used for purposes of diagnosing neurological diseases [1,2,3,9,12]. During automatic classification of diseases the additional units use elements of neural networks. The vectors based on normalised diagnostic measures [3] are inputs of the units. The measurements describe a patient"s posture condition, his walk and overloads occurring on his feet. The Counter-Propagation (CP), two-layer network has been used in one of the automatic conclusion-making units. During CP network activity, we can see not only supervised but unsupervised learning processes as well. This is a characteristic feature of the CP network. The initial steps of the CP network learning process are very important, because the success of the network training process depends on them to a great extent. Therefore, a new method of weight vector initial values selection was proposed. The efficiency of the method was compared with classical methods. The results were very satisfactory. Owing to the proposed method, the time of the network training process as well as the mean-square error and the classification error was reduced. The research has been carried out using clinical cases of some neurological diseases: Parkinson"s Disease, left-lateral hemiparesis and right-lateral hemiparesis after ischemic stroke. The measurements, which were made on a control group of patients without any neurological diseases, were the reference for these diagnostic classes.

An approach to location of a region including the posterior and anterior cruciate ligament in the MR knee images has been developed. The proposed method of the PCL location in T1-weighted MR knee images is based on entropy (or energy) measure of fuzziness and fuzzy C-means (FCM) algorithm. Then, edges of a region of interest containing the ligament are found. The procedure has been tested on clinical T1- and T2- weighted MR knee images resulting in a 3D visualisation.