Piotr Zarychta

Body surface potential mapping for detection of myocardial infarct sites

Using the additional information from multi-lead body surface potential recordings we aimed to study ECG features to predict the extent of infarcted myocardium as part of the 2007 PhysioNet/Computers in Cardiology Challenge. We studied potential and QT maps through key stages of the ventricular cycle assessing the 2 training and 2 test cases. Clinical assessment of the ECGs was provided by three cardiologists. QRS axis was abnormal in training case 1. ST was elevated in training case 1 and test case 2. T wave axis was abnormal in training case 2 and test case 1. T wave axis was different to QRS axis in training case 1. Cardiologists agreed that training cases 1 and 2 were anterior and inferior infarctions respectively, while they considered both test cases to be normal variations. The maps, however, showed significant abnormalities in the test cases.

Features extraction in anterior and posterior cruciate ligaments analysis

The main aim of this research is finding the feature vectors of the anterior and posterior cruciate ligaments (ACL and PCL). These feature vectors have to clearly define the ligaments structure and make it easier to diagnose them. Extraction of feature vectors is obtained by analysis of both anterior and posterior cruciate ligaments. This procedure is performed after the extraction process of both ligaments. In the first stage in order to reduce the area of analysis a region of interest including cruciate ligaments (CL) is outlined in order to reduce the area of analysis. In this case, the fuzzy C-means algorithm with median modification helping to reduce blurred edges has been implemented. After finding the region of interest (ROI), the fuzzy connectedness procedure is performed. This procedure permits to extract the anterior and posterior cruciate ligament structures. In the last stage, on the basis of the extracted anterior and posterior cruciate ligament structures, 3-dimensional models of the anterior and posterior cruciate ligament are built and the feature vectors created. This methodology has been implemented in MATLAB and tested on clinical T1-weighted magnetic resonance imaging (MRI) slices of the knee joint. The 3D display is based on the Visualization Toolkit (VTK).

Wearable System for Activity Monitoring of the Elderly

The paper presents a concept of an Ambient Assisted Living system for daily monitoring of the elderly. Description and testing of the prototype of a mobile data acquisition device for physical activity detection using inertial sensors is made. Testing has been performed on a mechanical test stand and a group of healthy individuals during normal walking. The purpose has been to assess the repeatability of acquired data. Obtained results are satisfactory and further development takes place.

Computer assisted location of the lower limb mechanical axis

The main aims of this research are the accuracy check control and ascertaining quality of a new method of knee joint alloarthroplasty. This method has two basic advantages. The first, the prosthesis location precision in the knee joint increases, secondary, the surgical procedure time shortens down to 45−60 min. As a result the infection susceptibility decreases.The first step of the presented method is finding the mechanical axis of the lower limb. In the second step, based on the analysis of each slice of a whole computer tomography series, a radiologist or an orthopedist has to qualify the pathology of bone structures within the knee joint. In this way the section planes within femur and tibia are determined. The third step is based on the three-dimensional model of femur and tibia. In order to extract bone structures from the computer tomography slices of the knee joint, the fuzzy C-means algorithm with median modification has been implemented. These images allow the three-dimensional structure of femur and tibia to be built. In the next step femur and tibia imprints have been created and the 3D models of both bones have been built. On the basis of the computer tomography slices of the knee joint the doctor has to indicate two healthy parts of both bones. Together with the created 3D models they permit building a patient-specific band with two gully holes. The gully holes are dedicated to Kirschner tool location marks. This special-individual band is apposed to selected bone structures of the knee joint during the surgery (knee alloarthroplasty).

Cruciate Ligaments of the Knee Joint in the Computer Analysis

The main aim of this research is the feature vector of the cruciate ligaments finding. This feature vector has to clearly define the ligaments structure and make easier diagnose their. The feature vector finding is based on the successive steps in extraction 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 T1-weighted (T1W) MRI 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 the feature vector is built.

ACL and PCL of the knee joint in the computer diagnostics

The essential aim of the presented studies is the feature vector of the anterior and posterior cruciate ligaments (ACL and PCL) finding. The feature vector must satisfy mandatory two tasks. The first one has to clearly define the ligaments structure and the second has to make easier diagnose their. The feature vector finding is based on the successive steps in extraction process of both ACL and PCL. In the first stage a region of interest including cruciate ligaments (CLs) is outlined. The automatic method of location of the CL on the T1-weighted (T1W) MRI (magnetic resonance imaging) 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 the feature vector is built. The most important criterium in the CLs diagnosis are the surface area and shape CL structures. Therefore the feature vector has to include the surface area and skeleton of the segmented and extracted structures of the cruciate ligaments.

Anterior and Posterior Cruciate Ligament – Extraction and 3D Visualization

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.

Posterior Cruciate Ligament — 3D Visualization

This paper shows a method which in the first step realizes an automatic registration of the Tl- and T2-weighted MR knee images, in the second step locates the posterior cruciate ligament (PCL) on the Tl-weighted MR knee images and in the third step permits 3D visualization of the PCL structures. The automatic registration process of the Tl- and T2-weighted MR knee images is based on the entropy (or energy) measures of fuzziness. A method of location of PCL on Tl-weighted MR knee images has been designed on the basis of both entropy (or energy) measure of fuzziness and fuzzy C-means (FCM) algorithm with median modification. The 3D visualization of the PCL structures procedure, which used registration and location steps, has been implemented in MatLab and converted to Visualization Toolkit (VTK).

Modelling of properties of the alloy tool steels after laser surface treatment

This paper presents the investigation results of the computer modelling of the surface layer hardness of hot work tool alloy steel alloyed with the ceramic powders using the High Power Diode Laser (HPDL). Laser treatment by remelting or alloying with the carbides was employed for improvement of the surface layer properties of tools made from the hot work tool steels. The developed model of the neural network make predicting possible for the surface layer hardness values. Further investigations should be concentrated on computer modelling of microhardness and resistance wear abrasion using the artificial neural networks. The surface layer of the hot work steel alloyed with ceramic powder using the HPDL have good properties and makes it possible for using in various technical and industrial applications. The artificial neural networks were used to determine the technological effect of laser alloying on hardness of the hot work tool steels. [Received 10 January 2007; Accepted 12 July 2007]

Automatic Registration Of The Medical Images - T1- And T2-weighted MR Knee Images

This article shows a new method of the automatic registration of T1- and T2-weighted MR knee images. This method is based on the entropy and energy measures of fuzziness and can be used in localization process of cruciate ligament. First, two sequences (T1- and T2-weighted) are converted to a fuzzy representation. Then, the entropy and energy measures are employed in the NCC (normalized cross correlation) and GD (gradient difference) methods. The alignment based on energy and entropy fuzzy measures shows a significant improvement in comparison with the implementation of the original image