Apnea is one of the deadliest diseases that can be prevented and cured if it is detected in time. In this paper, we propose a precise method for early detection of the obstructive sleep apnea (OSA) disease using the latest feature selection and extraction methods. The feature selection in this paper is based on the Dual tree complex wavelet (DT-CWT) coefficients of the ECG signals of several patients. The feature extraction from these coefficients is done using frequency and time techniques. The Feature selection is done using the spectral regression discriminant analysis (SRDA) algorithm and the classification is performed using the hybrid RBF network. A hybrid RBF neural network is introduced in this paper for detecting apnea that is much less computationally demanding than the previously presented SVM networks. Our findings showed a 3 percent improvement in the detection and at least a 30 percent reduction in the computational complexity in comparison with methods that have been presented recently.
Nonhuman primates are often used in biomedical research and to investigate physiologic processes that occur in man. Impedance plethysmography was used to measure calf, thigh, pelvic, abdominal, and thoracic volume changes in ten Rhesus and eight squirrel monkeys during five-minute exposures to HUT and HDT at angles of 5, 10, and 20 degrees. Calf, rump and tail measurements were made in three squirrel monkeys at 10 and 20 degrees of HUT and HDT. Fluid volume changes in all segments of the Rhesus monkeys were found to change during HUT an HDT in direct relation to the angle of tilt used. However, the volume changes that occurred in the squirrel monkeys were found to be quite different. Their calf, thigh, and pelvic segments lost volume during both HUT and HDT while their abdominal and thoracic segments responded similarly to those of the Rhesus monkeys. These results and those of the calf/tail measurements of the squirrel monkeys suggest that they may utilize their tails as a compensatory reservoir during postural changes and therefore, may not be an appropriate animal model for man under some orthostatic test conditions.
The present study aims to calculate a new database of conversion coefficients from fluence and air Kerma to personal dose equivalent in two terms: absorbed dose and Kerma-approximations. In this work, we propose a new equation to perform an analytical fit of our Monte Carlo (MC) calculated conversion coefficients for photons for different angles. Also, we have calculated the conversion coefficients using the EGSnrc code. The conversion coefficients have been calculated for beams of monoenergetic photons from 0.015 to 10 MeV, incident on phantom ICRU for angles of incidence from 0° to a 75° in steps of 15°. Our computed values agree well when compared with those published for the ICRU 57 in Kerma-approximations with statistical uncertainties in the calculation around 2%. We can conclude from this work that the analytical approach is successful and powerful such as Monte Carlo methods to calculate the operational quantities.
The purpose of this work was to develop and validate a Monte Carlo model for a Dual Source Computed Tomography (DSCT) scanner based on the Monte Carlo N-particle radiation transport computer code (MCNP5). The geometry of the Siemens Somatom Definition CT scanner was modeled, taking into consideration the x-ray spectrum, bowtie filter, collimator, and detector system. The accuracy of the simulation from the dosimetry point of view was tested by calculating the Computed Tomography Dose Index (CTDI) values. Furthermore, typical quality assurance phantoms were modeled in order to assess the imaging aspects of the simulation. Simulated projection data were processed, using the MATLAB software, in order to reconstruct slices, using a Filtered Back Projection algorithm. CTDI, image noise, CT-number linearity, spatial and low contrast resolution were calculated using the simulated test phantoms. The results were compared using several published values including IMPACT, NIST and actual measurements. Bowtie filter shapes are in agreement with those theoretically expected. Results show that low contrast and spatial resolution are comparable with expected ones, taking into consideration the relatively limited number of events used for the simulation. The differences between simulated and nominal CT-number values were small. The present attempt to simulate a DSCT scanner could provide a powerful tool for dose assessment and support the training of clinical scientists in the imaging performance characteristics of Computed Tomography scanners.
Aim: To conduct a study on the effect of random setup errors inpatient for dose delivery in Intensity Modulated Radiotherapy plans using Octavius 4D phantom.
Materials and methods: 11 patients with cancer of H&N were selected for this study. An IMRT plan was created for each patient. The IMRT quality assurance plans were transferred to Mosaiq workstation in a linear accelerator. These plans were delivered at the reference treatment position. Subsequently, the QA plans were delivered on the Octavius 4D phantom after introducing errors in various translational and rotational directions. The setup inaccuracies introduced varied from 1 mm to 5 mm along X, Y. These setup uncertainties were then introduced along X and Y direction simultaneously in equal measures. Similarly, IMRT plans were delivered also after introducing roll and yaw rotation of 1, 2 and 3 degrees in phantom. The deviation of gamma indices at all these positions was analyzed with respect to the reference setup position.
Results: The percentage of points passing the gamma acceptance criterion decrease as we increase the setup error. The change is found to be very insignificant with setup error up to 2 mm along X, Y or XY direction. Similarly, the rotational error of up to 3 degrees is found to be acceptable.
Conclusions: Small setup (< 2 mm) correction in patients may not adversely affect the dose delivery. But an error of similar magnitude in 2 directions simultaneously has a much greater impact on IMRT dose delivery.
In the present paper, some imaging properties of nanoparticles-based contrast agents including gold, bismuth, and silver were assessed and compared with conventional (iodinated) contrast agent in spectral computed tomography (CT). A spectral CT scanner with photon-counting detectors (PCD) and 6 energy bins was simulated using the Monte Carlo (MC) simulation method. The nanoparticles were designed with a diameter of 50 nm at concentrations of 2, 4, and 8 mg/ml. Water-filled cylindrical phantom was modeled with a diameter of 10 cm containing a hole with a diameter of 5 cm in its center, where was filled with contrast agents. The MC results were used to reconstruct images. Image reconstruction was accomplished with the filtered back-projection (FBP) method with hamming filter and linear interpolation method. CT number and contrast-to-noise ratio (CNR) of all studied contrast materials were calculated in spectral images. The simulations indicated that nanoparticle-based contrast agents have a higher CT number and CNR than the iodinated contrast agent at the same concentration and for all energy bins. In general, gold nanoparticles produced the highest CT number and CNR compared to silver and bismuth nanoparticles at the same concentration. However, at low energies (below 80 keV), silver nanoparticles performed similarly to gold nanoparticles and at high energies (120 keV), bismuth nanoparticles can be a good substitute for gold nanoparticles.
According to the Google I/O 2018 key notes, in future artificial intelligence, which also includes machine learning and deep learning, will mostly evolve in healthcare domain. As there are lots of subdomains which come under the category of healthcare domain, the proposed paper concentrates on one such domain, that is breast cancer and pneumonia. Today, just classifying the diseases is not enough. The system should also be able to classify a particular patient’s disease. Thus, this paper shines the light on the importance of multi spectral classification which means the collection of several monochrome images of the same scene. It can be proved to be an important process in the healthcare areas to know if a patient is suffering from a specific disease or not. The convolutional layer followed by the pooling layer is used for the feature extraction process and for the classification process; fully connected layers followed by the regression layer are used.
The imperative use of ionizing radiation in medicine causes the inevitable occupational exposure of the medical workers during the course of routine duties. The magnitude of health risk due to such radiation exposures has been described in terms of occupational radiation doses. In this context, it is obligatory to monitor, measure and document the radiation dose of occupationally exposed medical workers. This study aims to review the whole-body occupational radiation exposures of medical workers in Pakistan. Specifically, online literature published during 2000-2018 was reviewed for the occupational radiation exposures of Pakistani medical workers. Analysis of the extracted personal dosimetry data revealed that the total number of monitored medical occupational workers was 26046. The range of total cumulative and annual average effective doses was 94-15785 Person-mSv and 0.66-7.37 mSv, respectively. A significant number of the workers (25477; ~98%) received an annual dose below 5 mSv, while only 18 workers received an occupational exposure exceeding the annual dose limit of 20 mSv. It is expected that this study will provide a useful reference for evaluating and improving radiation protection and safety policies in the country.
In forests, edaphic microbial communities are involved in litter decomposition and soil forming processes, with major contribution to humification, especially bacteria and fungi being responsible for the main ecosystem services fulfilled by the soil. Research has been carried out aiming to characterize the structure and diversity of microbial communities in the Rendzic Leptosols (WRB) under natural deciduous forest from Visterna, Babadag Plateau and to assess their contribution to ecosystem services provided by soil. The paper presents the results of quantitative estimations and taxonomic composition of soil and litter communities of heterotrophic bacteria and fungi, identification of cellulolytic species, as well as the microbial biomass and global physiological activities expressed as soil respiration potential. More than a half of bacterial species were common in litter and soil (SI=57.14%) and were represented by dominant species of fluorescent or non-fluorescent pseudomonads and Bacillus subtilis but no similarity was found between the two fungal communities. Fungal populations included cosmopolitan species, such as antagonists and strong cellulolytic representatives of genera Penicillium, Trichoderma, Mortierella, Chaetomium, Epicoccum, Aspergillus. Microbial density and microbial biomass presented the highest values in the litter (684 mg C x kg-1 d.s.) and in surface horizon Am1 of soil profile than in the bottom layers. The highest diversity was found in Am1 horizon (0-5 cm) H’=1.983 bits and ε=0.869 for cellulolytic community. Soil respiration reflected the intense physiological activity of microbiome, with high values associated to numerous effectives of bacteria and fungi especially in surface horizon. Microorganisms identified contribute to formation of soil by recycling of nutrients, cellulose decomposition, the synthesis of stable organic matter (humic acids), aggregation of soil particles, biological control of pathogens by antagonistic activity. They improve plant uptake of water and nutrients by forming symbioses (ectomycorrhizae), thus modelling the structure of vegetation.