Information Technologies in Biomedicine

This paper presents a new hybrid fuzzy clustering method. In the proposed method, cluster prototypes are values that minimize the introduced generalized cost function. The proposed method can be considered as a generalization of fuzzy c–means (FCM) method as well as the fuzzy c–median (FCMed) clustering method. The generalization of the cluster cost function is made by applying the

L

p

norm. The values that minimize the proposed cost function have been chosen as the cluster prototypes. The weighted myriad is a special case of the cluster prototype when the

L

p

norm is the

L

2

norm. The cluster prototypes are the weighted meridians for the

L

1

norm. Artificial data set is used to demonstrate the performance of proposed method, and the obtained results are compared to the results from the FCM method.

Methods based on kernel density estimation have been successfully applied for various data mining tasks. Their natural interpretation together with suitable properties make them an attractive tool among others in clustering problems. In this paper, the Complete Gradient Clustering Algorithm has been used to investigate a real data set of grains. The wheat varieties, Kama, Rosa and Canadian, characterized by measurements of main grain geometric features obtained by X-ray technique, have been analyzed. The proposed algorithm is expected to be an effective tool for recognizing wheat varieties. A comparison between the clustering results obtained from this method and the classical

k

-means clustering algorithm shows positive practical features of the Complete Gradient Clustering Algorithm.

In this paper a new method of automatic image segmentation is presented. In proposed approach a region merging method is applied, where pixels or pixel groups are merged to bigger areas. In the proposed method of the region merging seed settings, segmentation parameter definition, and number of outcoming areas determining can be ommited. Homogeneous clusters (consisting of continuous groups of pixels) are the part of the growth process. The clusters are merged when the homogeneousness condition is fulfilled. The threshold value changes during segmentation process, appropriate to the changeable conditions.

In many biomedical fields there is a need of digital image analysis. The images often contain some disturbances in addition to the useful data. These disturbances should be reduces (or even eliminated, if it is possible) in order to improve the quality of the analysis. One of the possible methods of noise attenuation is low-pass filtering such as arithmetic mean and its generalization, namely weighted mean filtering where the weights are tuned by some adaptive algorithm.

This paper presents application of Bayesian weighted averaging to digital filtering 2D images which is some modification of the existing empirical Bayesian weighted averaging method created originally for noise reduction in electrocardiographic signal. The description of the new filtering method and a few results of its application are also presented with comparison to traditional arithmetic average filtering and median filtering.

There is a rapidly growing interest in electronic atlases including images with medical reports and quantitative description of pathological regions of interest. The e-atlases are mostly used by students and residents for teaching and training. However, an atlas my also be employed in everyday clinical practice. An imaging system that integrates an e-atlas of photodynamic studies with computer assisted diagnostic and therapeutic procedures is presented. Image processing results are stored in the database and may also be applied in the teaching and training process. The e-atlas includes photodynamic studies with clinical reports, location of the biopsy enter, selected points of the highest autofluorescence intensity to be subjected to the therapeutic light exposure.

The paper deals with the segmentation of brain tumours in magnetic resonance (MR) images. The segmentation method developed is based on the analysis of two MR series, and namely contrast enhanced T1W images and perfusion maps. The contrast enhanced T1W images are segmented with the use of the Kernelised Weighted C-Means (KWCM) method, yielding a binary mask of the suspected tumour. Next, the mask is superimposed over the perfusion maps. The perfusion intensities of corresponding areas in both brain hemispheres are compared. The results obtained for three subjects with glioblastoma multiforme (GBM) and anaplastic astrocytoma (AA) are discussed.

In this paper we present algorithm which purpose is to estimate real symmetry plane of the brain on the MRI perfusion studies. Because perfusion slices are acquired in low resolution and larger distances, than standard MRstudies, symmetry plane of the brain must be evaluated on the slice-by-slice basis. Presented algorithm therefore searches for symmetry line taking into consideration only anatomical information on each slice, ignoring adjacent slices. After presenting the method and sample results, we present results of tests.

Early detection of breast cancer is very important. It increases breast cancer treatment and reduces mortality rates by 30 - 70%. Architectural distortion (AD) is one of the commonly missed signs of breast cancer. It is estimated that 12 - 45% of missed breast cancer in mammography are ADs. Our ultimate goal is to develop a CAD module through creating of ADs detection method (

ArDist method

). We rely on fact that AD is a group of line structures of different orientation. The

ArDist method

consists of two stages: detection of ROI with potential ADs based on analysis with Gabor filters (

GF method

) and recognition of ADs using 2D Fourier transform in polar coordinates (

DD method

). The method was tested with 33 mammograms containing ADs, from the database

DDSM

. Experimental results are promising in comparison with the results of the model method and the efficiency of commercial CAD systems. The sensitivity of

ArDist method

amounts to 68% with 0.86 false positives per image.

A time resolved method based on near infrared spectroscopy for monitoring of changes in tissue absorption and concentration of the exogenoeus chromophore concentration is presented. We constructed a setup which can be used to measure changes in oxygenation of the brain tissue by monitoring of diffuse reflectance at two wavelengths. Currently, the instrument has been modified in such a way that simultaneous measurements of diffuse reflectance and fluorescence can be carried out. We present results of measurement of diffuse reflectance and fluorescence during injection of an optical contrast agent in healthy volunteer.

This paper presents the first stage of a semi-automatic method for the segmentation of nodules of the skeleto-muscular system from magnetic resonance (MR) imaging. Themethod suggested is efficient irrespective of the tumour location in human body. It is based on Fuzzy C-Means clustering (FCM), Gaussian Mixture Models (GMM) and Fuzzy Connectedness analysis applied to the dataset consisting of T1W and T2W series. In this study a method of transforming the results between planes is also presented. The suggested algorithm has been evaluated on the examinations of different parts of the body, where Ewing’s sarcomas have been indicated by a radiologist.

In order to obtain supporting tool for the pathologists who are investigating prognostic factors in folicular lymphoma a new method of color images segmentation is proposed. The method works on images acquired from immunohistochemically stained thin tissue sections of lymph nodes coming from patients with folicular lymphoma diagnosis. The proposed method of segmentation consists of: pre-processing, adaptive threshold, watershed segmentation and post-processing. The method is tested on a set of 50 images. Its results are compared with results of manual counting. It has been found that difference between the traditional method results and the proposed method is small for images with up to 100 nuclei while in more complicated images with more then 100 nuclei and with nuclei clusters this difference increases.

This paper shows successive steps in extraction and 3D visualization process of both anterior and posterior cruciate ligaments. In the first stage a region of interest including cruciate ligaments (CL) is outlined. The automatic method of location of the CL on the T1W MR knee images is based on fuzzy C-means (FCM) algorithm with median modification. The next step of that process is an extraction of the cruciate ligament structure using the fuzzy connectedness approach. In the last stage a 3D structures of the anterior cruciate ligament and posterior cruciate ligament are built. These spatial structure is created through layer composition of the base images by using a linear interpolation.

Assessment of bone disproportion is one of the routine radiological procedures frequently performed at a skeletal dysplasia diagnosis center. The existing knowledge on skeletal dysplasia patterns is more qualitative than quantitative, and the commonly applied diagnostic procedure is the manual, atlas-based evaluation of radiographs (including the hand) by an expert. A computer-aided diagnosis package (CAD) for bone age assessment was improved to analyze hand radiographs of skeletal dysplasia subjects with manifested hand bone abnormalities. The CAD software can automatically measure the lengths of phalangeal bones, compare them with those obtained from normal subjects collected in the Digital Hand Atlas (DHA), and recognize selected abnormality patterns. We have tested our method based on brachydactyly type A cases from the skeletal dysplasia diagnosis center that were matched with subjects from DHA in the age range of 2-12 years. New quantitative results obtained from this study have the potential to enrich the qualitative criteria currently utilized in the manual evaluation of abnormal bones, and thereby improve the performance of the diagnostic procedure.

A fully automated method for extracting brain structures from computed tomography images by employing adaptive filtering and finite Gaussian Mixture Modeling (GMM) with context-based enhancement is proposed. Generally, the method is composed of two phases. First, adaptive partial mean filter for noise removal and edge sharpening is used. The second phase is the multistage segmentation. Initial segmentation step concerning brain extraction from skull and non-brain tissue defines a region of interest (ROI) for further processing. Each pixel in ROI is assigned to one of three semantically fundamental classes - white matter (WM), gray matter (GM) and cerebrospinal fluid (CBF) and two extended classes of specific tissue. GMM with expectation-maximization algorithm (EM) is employed to assign initial class labels to image pixels and followed by context information modeling through Contextual Bayesian Relaxation Labeling (CBRL). The CBRL algorithm incorporates local neighborhood information and iteratively refines the outcome of GMM classification. The results of proposed approach have been verified by extracting susceptible-to-stroke regions (SSR) processed for hypodensity distribution estimation. The extracted structures are more smooth and reliable in comparison to region growing segmentation results.

Content-Based Image Retrieval (CBIR) is a field of rising interest for the application in Computer-Aided Diagnosis (CADx). Exploiting the visual information hidden in the images for retrieval, it facilitates the identification of similar past examinations, thereby providing a second opinion. Still, CBIR is not an integral part of a radiologist’s daily work. A comprehensive representation of CBIR results in a standard format utilizing established clinical infrastructure together with all referenced comparable examinations will support a bridge from CBIR to CADx. In this paper, we introduce the general IRMACON viewer, a system that represents results from CBIR systems encoded as a DICOM Structured Reporting document in a layout adjusted for CADx. IRMACON also allows the insight into identified similar examinations for convenient comparison by accessing patient’s information and diagnostic findings from the Hospital Information System (HIS) using HL7 messages. Our system is embedded into the clinical setting and the radiologist’s workflow. We applied the IRMACON viewer to the Image Retrieval in Medical Applications (IRMA) framework.

Features of trunk and postural deformations are clinically observed to detect related musculoskeletal disorders, namely scoliosis, kyphosis or other thoracic deformities. High variability and controversial reliability of results of postural screening require new solutions to be employed. Standardized criteria for correct body posture are needed for further studies. The aim of this work was to present the current status of development of the 3D diagnostic system for anatomical structures detection. The designed system is based on a parameterized method of body surface analysis for automated measurement of the three-dimensional shape of patient’s trunk. This system uses structured light to measure patient’s body in 3D space by projection of a set of raster images on its surface. A unique feature of the presented system is the ability to operate on remote sites and carry out measurement interpretation by telemedicine utilizing a data warehouse. The system comprises of three independent modules which deal with measurement, data archiving and its analysis. Communication between modules is performed over the TCP/IP protocol, incorporating two different channels for each client, dedicated to two kinds of data - text channel, for transmission of XML documents used for delivering commands and information and binary channel for the transfer of large binary data, such as clouds of points or photographs. Presented modules altogether allow to measure the full shape of subject’s 3D surface along with measurement data storage and analysis for support of screening and diagnosis.

Up until now the reconstruction of the pelvis defects after bone tumor resection has typically been by autologous or allogenous grafts. These approaches were highly unsatisfied because of large shape differences between the harvested transplant and the site to be reconstructed. There exists a huge demand for patient-specific and anatomically shaped implants. In this paper we propose a new pelvis reconstruction planning approach based on the statistical shape modeling. For generation of the statistical pelvis shape model a large data pool of CT datasets has been collected. Afterwards, the CT data segmentation and surface processing methods delivered the required pelvis geometries. Via Procrustes analysis of the collected pelvis surfaces the parameterized pelvis shape mean model has been calculated and the principal component analysis (PCA) [4] applied for estimating the anatomically optimal graft or implant geometry. In this work we demonstrate on a clinical pelvis reconstruction case that the using of statistical shape models in the oncologic surgery planning is a robust and very promising method. A quantitative evaluation of the matching quality and the convergence process is given.

The article proposes a new automatic method of the iridocorneal angle analysis in the anterior segment of the eye, using an original algorithm proposed by the authors. Results of analysis of a several tens of thousands images are shown in the later part of the article. There is also a proposal of a new method of measuring the iridocorneal angle, based directly on the image analysis and image processing methods. The algorithms have been implemented in the Matlab computing environment.

An increasing number of conducted endoscopic procedures results in inventing new computer systems, that would be a helpful tool in this kind of medical examinations. In this paper the need of evaluating algorithms describing the deformation of internal organs for laparoscopic surgery purposes is presented. A possible solution to the problem of visualising such deformations has been suggested. Furthermore, the way of collecting data by studying the liver reaction to an external force was described. A model of the liver deformations based on Bézier curves has been developed and compared with the shape of the real liver deformation.

The irregularity or ”roughness” of medical images is quantified by means of fractal dimension

D

. For medical images diagnostically important information often lies in the texture. In this paper we describe the application of the intensity difference scaling method for assessment of the fractal dimension

D

in the irregular regions of interest (irregular ROI-s). Near boundary between different tissues or structures the values of fractal dimensions changed significantly. It is difficult to fit entire regular region of interest within the examined organ with simultaneous inclusion of the relevant fragment, and at the same time to avoid the influence of boundaries.

Despite a great advancements in brain tumor diagnosis and surgery, a precise intraoperative location and removal of the neoplasmatic tissue remains difficult in many patients. For these reasons different navigation systems were developed. For the last several years neuronavigation has become widely used in brain tumor surgery. Surgical removal remains difficult in patients with small and deep seated brain lesions, multiple tumors, or location in important functional areas of the brain. Neurosurgeon faces the problem of causing severe neurological deficit as well as can have problems with finding and removing the lesion. Use of key hole procedures and surgical microscopes minimizes an operation field what forces to have a perfect orientation in the operation area, as well as precise data about the actual position of the instruments and their distance from the lesion. GOAL AND PURPOSE: Basing on their own surgical experience of 30 patients with brain tumors, the authors show the advantages and limitations of neuronavigation in surgical removal of brain tumors. MATERIAL AND METHODS: 30 patients with brain tumors located deeply in the hemispheres, skull and posterior fossa were analyzed. All patients were treated surgically with the use of neuronavigation equipment (Stryker Intellect Cranial Navigation System). Microsurgical technique with the use of magnification was introduced in all patients. After surgery, controlled CT examinations were performed in order to check the extension of tumor removal. RESULTS: Total surgical removal had been achieved in all patients. In all but one clinical result was excellent, the last patient died 2 weeks after removal of giant hypophyseal adenoma because of heart failure. CONCLUSIONS: Neuronavigation is an easy to use system which brings efficiency to surgical procedures. It allows to get a proper orientation in the operation field resulting in safe and accurate removal of intracranial neoplasms.

Final product of computer aided dental implants placements is an object created by means of Rapid Prototyping (RP) – surgical template. An essential condition that makes utilization of this special technique possible is computer tomography (CT), which has to be carried out first. Radiation might be in some cases inadvisable, not possible or not necessary. In this work presented are possibilities of using standard CAD techniques for the purposes of planning and assisting of dental implantology treatment based on introduced into a CAD software prosthetic foundation topography and a standard panoramic radiograph.

A proposition of the human’s reaction measurement system focused on the voice is presented in the paper. The main idea of this system is to test the human behavior under the influence of sound and visual stimulations. In order to evaluate the human reaction, the signal of speech is examined. This signal is acquired in strict conditions using external device which have a high measurement accuracy.

The article presents some behavioural measurements, together with the results of some biomedical research experiments and the assumptions underlying them. In order to determine the behavioural characteristics driving style was analysed, which was estimated using data recorded by a tachometric device. Biomedical measurements were made with using an ECG holter and infrared camera to record heart rate changes and the image inside the cab of the vehicle. On the basis of the results obtained, the assumption of a biomedically aided driver safety system is presented.

The myriad filter is a robust filter which is very useful in suppressing impulsive noise. It belongs to the family of the robust m-filters and is controlled with one parameter only. The myriad filter is defined as a running window filter whose output is the sample myriad of elements in the window. In this paper, a fast and simple method of the myriad filter computation is presented. It performs the 2nd order polynomial fitting and the next

x

-coordinate of the parabola top is searched. The proposed method can operate on different types of impulsive noise, requires less computational time and is equally robust as the fixed point method or the branch-and-bound search method. The presented method is applied to process a chirp signal in the impulsive environment. The obtained results are compared to the fixed-point algorithm and the branch-and-bound searching method.

This paper proposes an electrophysiological wireless patient monitoring system which integrates a Wireless ECG signal transmitter, GPS device and a mobile phone to acquire physiological signals and transmit them to a local server via Bluetooth wireless technology. Four kinds of monitor units were specially designed for a wireless communication, including a control center , a local monitor unit, mobile devices (personal digital assistant; PDA), and a Web page (for both patient and doctor). The use of various monitor units is created to fulfill different medical personnel requirements and wishes. This application was developed to promote the mobility and flexibility for the patients and also for the medical personnel, which further will improve both the quality of health care and lifestyle of the patient.

Major adverse cardiac events (MACE) are referred to as unsuspected heart attacks that include death, myocardial infarction and target lesion revascularization. Feature extraction and classification methods for such cardiac events are useful tools that can be applied for biomarker discovery to allow preventive treatment and healthy-life maintenance. In this study we present an entropy-based analysis of the complexity of MACE-related mass spectrometry signals, and an effective model for classifying MACE and control complexity-based features. In particular, the geostatistical entropy is analytically rigorous and can provide better information about the predictability of this type of MACE data than other entropy-based methods for complexity analysis of biosignals. Information on the complexity of this type of time-series data can expand our knowledge about the dynamical behavior of a cardiac model and be useful as a novel feature for early prediction.

The primary diagnostic tools allowing the detection of cardiovascular system diseases through the study of bioelectric activity of the heart are the electrocardiographic systems. The low amplitude ECG signal is usually disturbed by various types of noise, particularly by power line interference, which should be reduced without distortion of useful signal component. Since the amplitude (sometimes also frequency) of such noise varies in time, the adaptive approach may be applied to construct time-varying frequency characteristic filter.

This paper presents an approach to adaptive filtering based on set of infinite impulse response (IIR) bandpass filters, which responses are base for noise amplitude and frequency estimation. These values are used to adjust parameters of notch filter originally proposed by Mortara. The performance of the new method is numerically evaluated using the artificial ECG signal from CTS database combined with simulated power line noise with varying amplitude and frequency.

This paper presents a novel method for synchronous recording and analyzing both the electrogastrographic signal (EGG) and the heart rate variability signal (HRV). The four channel signals are non-invasively captured by six electrodes placed on the surface of the stomach. The EGG and electrocardiographic (ECG) signals, recorded simultaneously using the same electrodes and an amplifier, are separated by digital filtration. In our work the EGG and ECG analysis is limited to the calculation the most frequently used parameters: the dominant frequency (DF), dominant power (DP) of the power spectrum for the EGG, and LF/HF ratio of the HRV power spectrum for the ECG signals.

Estimation of electrooculography (EOG) and blinking signals could be done by using numerous algorithms. The separation of EOG and blinking signals is hard if the blinking pulses occur in the neighborhood of saccades. The appropriate estimation and signals separation could be obtained by using the optimization algorithm and a proper signals model. The proposed method improves the convergence of evolution–based technique and reduces the optimization time by using the additional algorithm for the estimation of possible positions of blinking pulses, the saccades positions and the EOG levels. A median filter based estimator is used for the initialization of optimization algorithm of blinking pulses. A differential filter based detector is used for the initialization of saccades position and the EOG signal level values between them. The proposed method reduces the computation time a few times what is important for the possibility of real–time implementation.

The article includes information about the advantages of Random Forests in DNA microarray data classification. The experiment presented as a background for the publication was performed on the data devoted to Barrett’s Esophagus and two types of Reflux Disease - Erosive and Nonerosive. An original idea of estimation of a quality of the classification evolved during studies on the problem and resulted in many interesting conclusions. There are presented topics such as advantages of Random Forests in supervised DNA microarray analysis, application of bootstrap resampling used for calculation of average quality results and comparison of classification quality for Random Forests, Support Vector Machines and Linear Discriminant Analysis. Proposed solutions are said to be a good measure of quality of classification with Random Forests method.

The main thrust of this work is to develop a bioinformatics tool for the automated evaluation of cell population homogeneity, which can be implemented on high content screening platforms. For that we employed Kullback-Leibler (K-L) divergence as a measure of similarity of multidimensional DNA codistribution patterns extracted from cell nuclei, and assigned analyzed cells into four categories. The resulting cell population homogeneity is determined as the ratios of cell quantities in the respective similarity categories. We evaluated our approach on human prostate cancer cells treated with an anticancer drug in comparison to untreated cells. A difference in homogeneities measured in these two populations was influenced by strong changes induced into DNA distribution patterns by treatment process vs. the occurrence of naturally more diverse groups of DNA patterns in the naïve cells.

Reliable segmentation of cell nuclei from microscopic images is an important task in many medical studies. This paper present a novel method for the segmentation of cell nuclei from microscopic images based on fuzzy clustering and ellipse fitting algorithm. It was designed specifically to segment nuclei in breast cancer FNB (Fine Needle Biopsy) cytological images. The segmentation approach takes both the color information and the spatial information into account during detecting the cell nucleus. Because of the presence of the elliptic shape function dissimilarity measure is able to differentiate the pixels with similar color but located in in different regions of the image. Moreover, it is very easy to extract morphometric parameters of the cell nuclei based on parameters of detected ellipses. Simulations and experimental results are provided to demonstrate the performance and generality of the proposed method. The proposed algorithm is able to segment cell nuclei obtained from microscopy imaging with reasonable accuracy.

Mathematical and physical principals of optical sectioning by using structured light in a conventional microscope are expanded. Assumptions and structure of a model of optical sectioning by using structured light are presented. The results of this model are discussed and compared with a reference method.

The novel computer aided image processing analysis presented in this paper is used to evaluate the number of bacteria colonies. Proposed algorithm exploits the computer-aided calculations of optical transforms and it is implemented to the analysis of CCD recorded images of bacteria colonies. The diffraction patterns were approximated by the Fourier spectrum and additionally, the Mellin transform was applied to omit the problem of bacteria colonies sizes fluctuations. This approach exploits the properties of Fourier and Mellin transforms for scale and spatial locations invariant analysis. The significant correlation between parameters characterizing the Mellin spectrum and the number of bacteria colonies, was observed. Presented method was used for characterization of sterilization efficiency of antibacterial nanomaterials.

This paper addresses principles of discrete nonlinear control performed towards the diagnostic quality-based interpretive software optimization for cardiology-based monitoring system. Thanks to the pervasive access to wireless digital communication, distributed biosignal interpretation networks may not only optimally solve difficult medical cases, but also adapt the data acquisition, interpretation and transmission to the variable patient’s status and availability of technical resources. The adaptation is based on the innovative use of results from the automatic ECG analysis to the seamless remote modification of the interpreting software. This paper focuses on the static properties and control accuracy and dynamic properties of the ECG processing optimization. It provides also details on the automatic scoring of the performance of ECG interpretation procedures. Testing of the prototype, despite its limited scale, yields significant quantitative benefits and improvement of diagnostic quality.

The influence of sleep conditions to human health and performance is currently well known but still underestimated and monitoring devices are not widespread. This paper describes measurement methodology and prototype design of a home-care sleep scoring device. The proposed solution is oriented towards low-cost equipment and easy-to-use data capture using contactless recording as far as possible. Unlike the regular laboratory systems, the quality of sleep is estimated from the video-recorded subject motion, audio-recorded acoustic effects and from the single-lead electrocardiogram being the only electrical signal recorded from the body surface. The presented prototype is built of consumer-grade devices working in a short-distance network and providing multimodal data. The information provided from different modes are partly redundant, giving opportunity for refinement of the accuracy, and partly complementary, widening the aspect of sleep analysis.

Surveillance systems are currently the most developed branch of assisted living applications providing the disabled or elderly people with unprecedented security in their independent life. This paper presents a design of a telemedical surveillance system, where graph theory is used to describe subjects’ states. Patient’s states expressed by sets of medically-derived parameters and his or her daily activity (a behavioral pattern) are represented by attributed probabilistic graphs with indexed and labeled nodes. This representation provides high flexibility in a state and transient description as well as a reliable measure of behavior divergence, which is a basis for automatic alerting. The system is designed for the subject’s apartment and supports a localization-dependent definition of his or her usual and unusual behavior. The apartment topology is also represented in the form of a graph determining subject’s pathways and states. This approach has been found very flexible in all aspects of personalization, appropriate to work with the behavioral presumption set or with the auto-adaptive artificial intelligence recognition engine. Also the patient’s state, thanks to the semantic description may be easily extended or refined if necessary by adding new, complementary data capture methods.

This paper presents the data capture methodology and design of a home care system for medical-based surveillance and man-machine communication. The proposed system consists of the video-based subject positioning, monitoring of the heart and brain electrical activity and eye tracking. The multimodal data are automatically interpreted and translated to tokens representing subject’s status or command. The circadian repetitive status time series (behavioral patterns) are a background for learning of the subject’s habits and for automatic detection of unusual behavior or emergency. Due to mutual compatibility of methods and data redundancy, the use of unified status description vouches for high reliability of the recognition despite the use of simplified measurements methods. This surveillance system is designed for everyday use in home care, by disabled or elderly people.

Open standards and ”open source” software are seen as those which are largely contribute to solving the problems of integration and interoperability in the newly-built computer information systems. In this work are presented views of the authors who propose certain conceptions and ideas about ICT in medical area based on the authors’ opinions, observations and the need of introducing changes in the healthcare. The perception of XML as the information exchange format, evolution of the exchange systems and data processing in the context of building Electronic Platform to Collect, Analyze and Sharing Digital Resources about Medical Events are main issues in this work.

The paper presents the results of kinematic analysis of the prototype of rehabilitation device for human upper limb. The aim of these measurements was to assess possibility to execute PNF movements in upper limb joints and correctness of functioning of control algorithms responsible for reproducing of assigned movements The measuring site consisted of a set of 4 digital cameras, computer workstation, a set of markers, calibrating dice and light sources. Measurements were carried out for simple and complex movements. First a man forced a motion sequence, and then the device had to repeat it. Measurements revealed necessity of changes in constructions due to unacceptable positioning inaccuracy. Exemplary results before and after these changes are presented in the paper.

Limb disability, which in many cases results from disease or injury, impairs everyday actions and limits job opportunities. Treatment consists of physical therapy that requires laborious repetition of recommended motions with the help of an experienced physiotherapist. Robot-aided exercises for repeatable rehabilitation motions would allow speeding up the rehabilitation process and reduce its costs. This work presents a concept of a control system for a rehabilitation robot constructed for this aim. The solution has been verified with the prototype robot ARM-100 for upper limb rehabilitation with the control system based on the assumptions discussed in this paper.

The aim of the study was to take advantage of a three-dimensional minimal spanning tree (3DMST) to model the crack process of materials, based on an example of a metallic biomaterial. The modelling of the crack process and its characterization were carried out on a stereometric measuring file acquired as a result of surface analyses of the biomaterial examined, using confocal microscopy.

Analysis of issues related to the crack process requires a quantitative description of the problem which frequently, due to its complexity, is difficult or impossible to solve. In a number of cases, the deficiencies of a quantitative description made using classical methods are compensated for by unconventional mathematical tools, such as the multifractal analysis, which was carried out for the obtained model of the crack path.

The main purpose of this paper is to show possible positive effect of 3D modeling application in rehabilitation. Authors provide clinic example of patient with ACL reconstruction. The patient has undertaken the therapy consisting of regular rehabilitation sessions as well as those enhanced by 3D rehabilitation application. The improvement in ACL rehabilitation was significant and worth further exploration.

The numerical analysis of spine stabilizers was carried out in the work. The aim of the work was determination of more favourable stabilization variant. Moreover, this analysis is the basis to determine an optimal geometry of the stabilizer as well as to select mechanical properties of metallic biomaterial. The work presents the results of biomechanical analysis of lumbar spine stabilization by plate and transpedicular stabilizer. The finite element method was applied for biomechanical characteristic of implants made of stainless steel Cr-Ni-Mo and Ti-6Al-4V alloy. Geometrical models of part of spine L3-L4 and the stabilizer were discretized by SOLID95 element. Appropriate boundary conditions imitating phenomena in real system with appropriate accuracy were established. The result of numerical analysis was calculation of displacements and reduced stresses in particular elements of system in a function of the applied loading: 700 - 1600 N. The displacements calculated for part L3-L4 show that the proposed type of stabilizer enables the correct treatment process. The obtained numerical results should be verified in ”in vitro” tests.

The aim of the work was biomechanical analysis of a low contact plate (patent no. P. 387934) applied in treatment of tibia. Shape of the plate was adapted to anatomical curvature of the bone. Biomechanical analyses were carried out for the implant made of two metallic biomaterials used in bone surgery - stainless steel and Ti–6Al–4V alloy. Finite element method was applied to calculate displacements and stresses. The obtained results allowed to work out biomechanical characteristics of the analyzed systems. These characteristics can be a basis for selection of strain hardening of the applied metallic biomaterials as well as the optimization of the plate’s geometry.

The fundamental purpose of this research was to determine the biomechanical characteristics of the first metatarsal bone - double-threaded screws system made of stainless steel (Cr-Ni-Mo) and an assessment of its stability. To define the biomechanical characteristics of the system, the finite element method and experimental method were applied. Geometric models of metatarsal bone and double-threaded screws, were discretized by means of SOLID 95 element. Appropriate boundary conditions imitating phenomena in the real system with appropriate accuracy were established. The aim of biomechanical analysis was calculation of displacements and stresses in the bone and the stabilizers in a function of the applied loading. The experimental method was carried out to calculate displacements of the analyzed system. The obtained results can be applied to determine the construction features of the stabilizer and to select mechanical properties of metallic biomaterial (selection of degree of strain hardening).

The main aim of this work is a biomechanical analysis of a surgical drill-femur system under conditions simulating the bone drilling process, using a finite-element method. The geometrical model of the surgical drill was prepared for different point angle values of the drill tip (2

κ

1

= 90

o

and 2

κ

2

= 120

o

) and different diameters (

d

1

= 9.0 mm,

d

2

= 4.5 mm,

d

3

= 3.2 mm and

d

4

= 1.0 mm). The analysis included two variants of the drilling process: Variant I included drilling a single layer of the cortical tissue of a femur, while Variant II included the simulation of another step of the drilling process - drilling in the opposite area of cortical tissue. Calculations were made for two types of drill material. The biomechanical analysis may form the basis for improving the geometry of surgical tool tips and optimising a selection of the mechanical properties of the material used to manufacture them.

This work describes the structural strength analysis of selected drills used for mechanical root canal preparation. The assumptions made in the biomechanical analyses included both the technique of performing the surgery and various shapes of root canal. The calculations were using two alternative types of metal materials used to manufacture surgical instruments - X39Cr13 martensitic steel and Ni-Ti alloy with super-elastic properties. The results of the analyses indicate that using Ni-Ti alloy in the production of endodontic drills assures safety during root canal treatment processes.

An upright posture is a dynamic motion process, the stability of which depends on how the centre of gravity is projected onto the base of support. One of the basic external manifestations of this function is the dimensional symmetry of vectorial, angular and amplitudinal motion determinants. The nervous system, in its reaction to pain or to developed anatomical alterations (e.g. the shortening of a limb), strives to correct the body structure symmetry and its dynamic parameters, hence the asymmetry determiner of a feature related to a foci of pain or dysfunction may be used indirectly and directly to achieve a clinical quantitative evaluation. The aim of the study was to perform a comparative evaluation of the symmetry of standard densitometric parameters of the pelvic joints in relation to determinants of the pelvis x1-6 and lower body limbs, as well as to the difference in length of the lower limbs and symmetry of the shoulder girdle muscle structure determiners y1-3 in selected pain conditions. Following a radiogram analysis of selected patient groups, a high correlation of y1-3 asymmetry determiners in relation to the difference in length between the lower limbs and to the x1-6 asymmetry determiners was observed. The study showed only a slight correlation with the asymmetry of densitometric parameters of the hips evaluated using a DXA absorption meter.

The bronchoscopy diagnostics consists of numerous tasks including but not limited to the treatment planning on the basis of computed tomography, registration and analysis of endoscopic images during the inspection of the tracheo-bronchial tree and procedure management. There is no comprehensive solution for a computer system supporting bronchoscopy laboratory. Since the limited time of medical personnel, high volume and multimodal data created during bronchoscopy procedure and complex workflow we decide to build a dedicated system addressing these issues. The paper presets an overview of the bronchoscope diagnostics, user needs and requirements related to the procedure and the proposed solution. Finely, the deployment results as well as future plan is presented.

In the paper, an advanced visualisation module of recently developed BRONCHOVID system designed for efficient data exploration and bronchoscopy guidance is presented. It brings an intuitive and user-friendly interface developed with several open source subroutines as well as diversified modes and algorithms of thoracic CT of data segmentation and visualisation including a surface shaded display (SSD), volume rendering and multiplanar reformatting. For enhanced concurrent visualisation of thoracic structures located in the vicinity of or adjacent to trachea and bronchial tree (lymph nodes, blood vessels, pathological masses) a new dedicated segmentation algorithm has been developed. It is based on a region growing method followed by the use of deformable models. The module integrates also the options of central path calculation by means of distant transform in lower airways and interactive virtual bronchoscopy. Its functionality covers also an option for measurements of dimensions in volume of CT data.

The video recordings of endoscopic procedures performed within respiratory tract include both frames of adequate and inadequate quality for the assessment by the endoscopist. The frames of inadequate quality were called by some authors blurred or “non-informative”. The fraction of blurred frames within video recording of bronchofiberoscopy may be considerable and it varies from case to case. Therefore, the function of automatic exclusion of “non-informative” frames would bring substantial benefits in terms of the volume of the archived video recordings of bronchofiberoscopic procedures. Furthermore, it could also save the time of users accessing medical video library established with archived resources. In this paper, the authors have proposed, tested and compared several algorithms for detecting blurred video frames. The main focus of this paper is to compare various, independent algorithms for automatic recognition of “non-informative” frames in video recordings of bronchoscopic procedures. The results demonstrated in the paper show that the proposed methods achieve F-measure, sensitivity, specificity and accuracy of at least 87% or higher.

A medical digital video library can contain numerous bronchoscopy video sequences. Searching such database is very difficult and can take lots of time of a medical specialist. Therefore, automatic searching algorithms should support such a task. In this paper we presented two models supporting automatic detection of bleeding and mucosal pallor. Both models are based on MPEG-7 descriptors and both were built with a scoring model. The presented ROC plots show that the obtained results can be very helpful in a real system.

Current problem of computer assistance of bronchoscopy neoplasmatic lesion interpretation is still open and exceptionally challenging question. Strongly diversified pathology appearance, lack of reference patterns, subjective criteria of lesion assessment and often unsatisfactory quality of registered video limit diagnostic effectiveness of bronchoscopic studies. Fundamental part of presented research is an analysis of various endobronchial tumor manifestations in a context of acquired image sequences, bronchoscope navigation, artifacts, lightening and reflections, changing color dominants, unstable focus conditions, etc. Representative patterns of pathologies in bronchoscopy images and standardized features of malignancy were sought. Proposed method of neoplasmatic areas indication was based on three fundamental steps of preprocessed video analysis: 1) informative frame selection, 2) block-based unsupervised determining of enlarged textural activity, 3) indication of the regions with potentially tumor signatures, based on feature selection in different domains of transformed image followed by SVM classification. Efficiency of pathology recognition was verified with a reference image dataset of 14 examinations containing diversified neoplasmatic endobronchial tumor patterns. Experimental results reveal promising accuracy of recognition for differential forms of endobronchial tumor manifestations. Test set of selected informative blocks was classified as pathological or normal with accuracy of 85%.

The paper presents results of SIFT (Scale-Invariant Feature Transform) method application to support transbronchial needle-aspiration biopsy. An image generated during a bronchoscopic examination is often disturbed by vomiting reflex of the patient, or by sudden influxes of saliva or mucus. The presented system supporting the doctor is capable of monitoring the examined spot and signalling return to this spot after the disturbance ends or after the repeated examination of the same fragment of bronchial tree. A modified SIFT method is used, with a particular focus on real-time SIFT calculation.

Intensive endoscopic examinations result in an increasing number of records in video format. Storing whole videos is ineffective when considering mass storage devices usage. Additionally, reviewing these videos is a very time-consuming task.

In the BRONCHOVID [1] project some methods of elimination of less informative video parts were proposed. The final effect was a very shortened registration which was called a

summarization

. A summarized video could be supplemented with additional information (data), important from the diagnostic or anatomic point of view. The first group consists of markings of biopsy sites and frames on the basis of which lesions were diagnosed. In the BRONCHOVID project, similarly as in other congruent projects, the aim is to detect pathologies automatically. Information obtained in this way should be verified by a doctor.

Individual bronchoscopy test event takes about 20 minutes. Most of this time is the video recording process, placing and removing bronchoscope endings in the patient’s airways. A sizable part of the record is unreadable because of obstruction by secretion of physiological image. The analysis of such recordings for teaching purposes or diagnostic tests is time-consuming — the doctor or student is forced to view a large number recordings of little value to find the fragments of interest — those showing lesions. In research, we develop a prototype system to create shortcuts (called summaries or abstracts) of bronchoscopy research recordings. Such a system, based on the model described in the preceding paper paragraphs uses image analysis methods to delete the recording fragments of the bronchoscopy test which have no diagnostic value for teaching and create a few minute-long, valuable video sequences.

The perception enhancement of bronchovideoscopy examination was the subject of reported study. Proposed methods enrich diagnosis and treatment giving real-time additional view based on following adjusted concepts: contrast enhancement, histogram equalization, sharpening, color correction. Additionally wavelets, edge preserving and morphological filters were involved in the image processing. Wavelet texture extraction and sharpening were considered as the most useful and included into complex system of bronchoscopy storage, review and analysis called “Bronchovid”. Extended capabilities of classical white light bronchoscopy which has higher resolution than other autofluorescence and spectroscopy based diagnostic methods can be complementary for them in case of any endobronchial biopsy suspicious findings.

The aim of the study performed in the Department of Biophysics of Nicolaus Copernicus University is to investigate the effect of extremely low frequency electromagnetic field (ELF-EMF) exposure on animal nervous system function. American cockroach (Periplaneta americana) serves as a model in experiments. Insects are placed into the chamber inside the coil (electromagnetic field exciter) designed and manufactured by ,,Elektronika i Elektromedycyna Sp. J.”. Before starting the exposure it is necessary to determine: (1) what kind of field is applied, (2) the parameters of exposure and (3) the area of homogenous field. Only well established set-up characteristics can provide correct and comparable conditions of exposure to EMF. As well as, before starting the observation of the EMF influence on insects it is necessary to exclude factors which are associated with EMF exposure (such as temperature changes and vibrations) and could affect the organism function (behavior of animal). Especially cockroaches are very sensitive to vibrations and mechanical disturbances. Thermal conditions during the whole period of EMF exposure have been established earlier 1. The research that has been conducted by the authors so far included methodology of measurements of mechanical vibration spectrum and magnetic field distribution in interior part of electromagnetic field exciter on insects putted into research chamber. The vibration of electromagnetic field exciter occurs due to current flow by the exciter windings 2,3, and they acts on it in radial and longitudinal direction. In this paper the vibration and field distribution of field exciter measurements results are presented.