Journal of Biomedical Physics & Engineering
Volume:3 Issue: 4, Jul-Aug 2013

Brain tissue segmentation for delineation of 3D anatomical structures from magnetic resonance (MR) images can be used for neuro-degenerative disorders, characterizing morphological differences between subjects based on volumetric analysis of gray matter (GM), white matter (WM) and cerebrospinal fluid (CSF), but only if the obtained segmentation results are correct. Due to image artifacts such as noise, low contrast and intensity non-uniformity, there are some classification errors in the results of image segmentation. An automated algorithm based on multi-layer perceptron neural networks (MLPNN) is presented for segmenting MR images. The system is to identify two tissues of WM and GM in human brain 2D structural MR images. A given 2D image is processed to enhance image intensity and to remove extra cerebral tissue. Thereafter, each pixel of the image under study is represented using 13 features (8 statistical and 5 non- statistical features) and is classified using a MLPNN into one of the three classes WM and GM or unknown. The developed MR image segmentation algorithm was evaluated using 20 real images. Training using only one image, the system showed robust performance when tested using the remaining 19 images. The average Jaccard similarity index and Dice similarity metric for the GM and WM tissues were estimated to be 75.7 %, 86.0% for GM, and 67.8% and 80.7%for WM, respectively. The obtained performances are encouraging and show that the presented method may assist with segmentation of 2D MR images especially where categorizing WM and GM is of interest.

The most common intravascular brachytherapy sources include 32P, 188Re, 106Rh and 90Sr/90Y. In this research, skin absorbed dose for different covering materials in dealing with these sources were evaluated and the best covering material for skin protection and reduction of absorbed dose by radiation staff was recognized and recommended. Four materials including polyethylene, cotton and two different kinds of plastic were proposed as skin covers and skin absorbed dose at different depths for each kind of the materials was calculated separately using the VARSKIN3 code. The results showed that for all sources, skin absorbed dose was minimized when using polyethylene. Considering this material as skin cover, maximum and minimum doses at skin surface were related to 90Sr/90Y and 106Rh, respectively. polyethylene was found the most effective cover in reducing skin dose and protecting the skin. Furthermore, proper agreement between the results of VARSKIN3 and other experimental measurements indicated that VRASKIN3 is a powerful tool for skin dose calculations when working with beta emitter sources. Therefore, it can be utilized in dealing with the issue of radiation protection.

Accelerated partial breast irradiation via interstitial balloon brachytherapy is a fast and effective treatment method for certain early stage breast cancers however skin, chest wall and Lung doses are correlated with toxicity in patients treated with breast brachytherapy. To investigate the percentage of the dose received by critical organ (skin), thermoluminescence detector was used in MammoSite brachytherpy and the ability to control skin dose between MammoSite and MultiCatheter brachytherapy was compared with each other. Dosimetry is carried out using a female-equivalent mathematical chest phantom and Ir-192 source for brachytherapy application. Our initial results has shown good agreement with surface doses between those calculated from the treatment planning results and those measured by the thermoluminescence detector. The mean skin dose for the experimental dosimetry in MammoSite was 2.3 Gy (56.76% of prescription dose). The results show that the MultiCatheter method is associated with significantly lower mean skin and chest wall dose than is the MammoSite. The Multi- Catheter technique is quite flexible and can be applied to any size of breast or lumpectomy cavity, But in MammoSite technique, verification of balloon symmetry, balloon/ cavity conformance and overlying skin thickness is essential to assure target coverage and toxicity avoidance.

Radon and its daughters are amongst the most important sources of natural exposure in the world. Soil is one of the significant sources of radon/thoron due to both radium and thorium so that the emanated thoron from it may cause increased uncertainties in radon measurements. Recently, a diffusion chamber has been designed and optimized for passive discriminative measurements of radon/thoron concentrations in soil. In order to evaluate the capability of the passive method, some comparative measurements (with active methods) have been performed. The method is based upon measurements by a diffusion chamber, including two Lexan polycarbonate SSNTDs, which can discriminate the emanated radon/ thorn from the soil by delay method. The comparative measurements have been done in ten selected points of HLNRA of Ramsar in Iran. The linear regression and correlation between the results of two methods have been studied. The results show that the radon concentrations are within the range of 12.1 to 165 kBq/m3 values. The correlation between the results of active and passive methods was measured by 0.99 value. As well, the thoron concentrations have been measured between 1.9 to 29.5 kBq/m3 values at the points. The sensitivity as well as the strong correlation with active measurements shows that the new low-cost passive method is appropriate for accurate seasonal measurements of radon and thoron concentration in soil.

The time and frequency features of motor unit action potentials (MUAPs) extracted from electromyographic (EMG) signal provide discriminative information for diagnosis and treatment of neuromuscular disorders. However, the results of conventional automatic diagnosis methods using MUAP features is not convincing yet. The main goal in designing a MUAP characterization system is obtaining high classification accuracy to be used in clinical decision system. For this aim, in this study, a robust classifier is proposed to improve MUAP classification performance in estimating the class label (myopathic, neuropathic and normal) of a given MUAP. The proposed scheme employs both time and time–frequency features of a MUAP along with an ensemble of support vector machines (SVMs) classifiers in hybrid serial/parallel architecture. Time domain features includes phase, turn, peak to peak amplitude, area, and duration of the MUAP. Time–frequency features are discrete wavelet transform coefficients of the MUAP. Evaluation results of the developed system using EMG signals of 23 subjects (7 with myopathic, 8 with neuropathic and 8 with no diseases) showed that the system estimated the class label of MUAPs extracted from these signals with average of accuracy of 91% which is at least 5% higher than the accuracy of two previously presented methods. Using different optimized subsets of features along with the presented hybrid classifier results in a classification accuracy that is encouraging to be used in clinical applications for MUAP characterization.