Search

Tongue segmentation is an essential part of intelligent medicine diagnosis. The purpose is to generate an accurate contour of the tongue region by a precise mask. In recent years, deep learning methods have been widely applied in the field of image processing and have achieved impressive performance. With the increasing requirement of the performance for medical image segmentation, many scholars have employed deep learning to tongue segmentation. The methods of deep learning-based tongue segmentation are analyzed, classified and summarized. In the field of tongue segmentation applications, various tongue segmentation methods based on deep learning are divided into eight types: convolutional neural network (CNN), fully convolutional network (FCN), convolutional model with graphical model, encoder-decoder based model, regional convolutional network-based model, atrous convolutional model, transfer learning and other methods. This paper presents a comprehensive survey of the recently developed deep learning for tongue segmentation, and analyzes the strengths and weaknesses of these methods. The commonly used data sets and evaluation indexes of tongue segmentation based on deep learning are visually compared and quantitatively evaluated. This survey is concluded by discussing the potential development in future research work.

The performance of gradient and shim coils affect the quality of magnetic resonance imaging (MRI) directly. Conventional gradient/shim coils have some inherent disadvantages, such as the state of produced magnetic field is single and inflexible, many types of coils are required and the structure is complex, while the matrix gradient coils are able to overcome those problems. This paper introduces the concept and characteristics of matrix gradient coil, and reviews the current progresses in this research topic according to the structure and function of the gradient coil. The development tendency of matrix gradient coil in the future is proposed. In addition, the preliminary research on matrix gradient coil is presented.

Epilepsy is a common neurological illness. By analyzing the electroencephalo-graph (EEG) signals of patients with epilepsy, their conditions can be monitored so that seizure can be detected and intervened in time. In epilepsy research, how to obtain effective features and construct an effective classifier is the key to epilepsy detection and recognition. In order to obtain better epilepsy detection effect, this paper proposes a multi-view seizure detection algorithm that combines deep and shallow features. The method first uses FFT (fast Fourier transform) and WPD (wavelet packet decomposition) to obtain the shallow features in the frequency domain and the time-frequency domain. Then, CNN (convolutional neural network) is used to obtain the deep features in the frequency domain and the time-frequency domain. Further the multi-view TSK fuzzy system is used to construct the classification model of shallow and deep features. Experimental studies show that in the detection of EEG signal epilepsy, the effects of shallow features and deep features proposed in this paper are more than 1% higher than those of commonly used feature extraction methods such as PCA (principal component analysis), LDA (linear discriminant analysis) and etc. The multi-view epileptic detection algorithm which combines deep features and shallow features has a classification effect higher than that of the single-view algorithm by more than 1%, and the average detection effect is more than 5% higher than that of the single-view algorithm.

Ethanol is a feasible medium for diffusion separation of carbon isotopes. In order to obtain the separation coefficient of gas diffusion separation of ethanol, the application of elemental analysis isotope ratio mass spectrometry (EA-IRMS) to measure the carbon isotope abundance in ethanol in gas diffusion separation experiment was studied. Through literature investigation, this study optimized the preparation process of ethanol samples, developed the method of EA-IRMS for the determination of carbon isotope abundance of ethanol, tested the stability, and realized the measurement of carbon isotope abundance of ethanol samples. Based on the separation experiment of gas diffusion method, the carbon isotope abundances of refined ethanol and lean ethanol in multiple separation experiments are obtained. The basic total separation coefficient of carbon isotopes by ethanol diffusion can be calculated by formula derivation. This study provides an analytical basis for the future experiment of diffusion separation of carbon isotopes with ethanol as medium.

Electromagnetic metamaterials have attracted more and more research interest due to their exotic physical properties and important applications in manipulating electromagnetic waves. In this paper, the research progress in terahertz metamaterials and their applications in imaging was reviewed: firstly, the research of terahertz metamaterials was overviewed, and tunable and reconfigurable terahertz metamaterials as well as terahertz digitally coding and field-programmable metamaterials were discussed in detail; then, the applications of terahertz metamaterials in imaging were illustrated, including terahertz imaging technologies based on metasurface lenses, metamaterial absorbers, reconfigurable metasurfaces, and field-programmable metasurfaces; finally, the prospects of terahertz metamaterials and their applications in imaging were provided.Reconfiguration and in telligence of the functions will be an important development direction of terahertz metamaterials,and the emerging information metamaterials which integrate meta meaterials and information technology will enable more efficient and convenient terahertz imaging.

Aiming at the shortcomings of the existing wired charging mode, such as complex wiring and lack of flexibility, a wireless charging transmission system for electric vehicles is designed. The electromagnetic and circuit parts are studied and designed respectively. The coil model of the wireless charging transmission system is obtained and the coil is manufactured successfully. The design coil is simulated by ANSYS Maxwell software. The inductance coefficient, mutual inductance coefficient and coupling coefficient between coils are obtained. The simulation results verify the feasibility of the design system.

Proton-heavy ion therapy is internationally recognized as the most advantageous radiotherapy method in oncology. However, due to the relatively high energy of the proton-heavy ion beams, the secondary neutrons produced in the process of beam energy loss activate the surrounding materials and generate induced radioactivity, which is a threat to the health of staff and patients. Therefore, it is necessary to pay attention to the related radiation protection. In this paper, the causes and distribution of induced radioactivity were explored. We estimated the energy distribution and radiation dosages of the secondary neutrons produced in proton-heavy ion therapy by performing measurements and MCNPX (Monte Carlo) simulations, and we also detected the radiation dosages of the induced radioactivity varying with time after irradiation. Finally, we discussed the effective radiation protection measures with the aim to reduce and avoid the health hazards from induced radioactivity to the staff and patients and provide practical guarantee for the safety and effectiveness of proton-heavy ion therapy.

Experimental validation of beam spectral source term parameters is one of the important procedures for boron neutron capture therapy (BNCT) physical dosimetry. It is crucial that the well-known the relevant information to precisely establish clinical treatment planning for correctly assessing the dose delivered to patients. To verify the reliability of the theoretically calculated photon energy spectra of source term, a new photon spectrometer applicable for BNCT beams (n/γ mixed beams with wide energy range and large radiation intensity) was designed. In order to realize the fast and accurate measurement of the photon spectrum for BNCT therapy beam, the radiation intensity should be reduced to prevent radiation damage and control the dead time of measuring system at a proper level, meanwhile, the neutron induced secondary photons must be suppressed as much as possible. The neutron fluence rate, photon count rate and secondary photon count rate proportion (secondary photon count rate/source term photon count rate) in the detector are optimized by Monte Carlo simulation, the radiation intensity at detector position is decreased largely, the secondary photon count rate proportion is reduced to 5.45. Furthermore, the calibration method of spectrometer response to different energy photons was also investigated at reference photon fields so as to obtain accurate response function for unfolding the photon spectra.

At present, the electron beam radiation processing industry has developed rapidly in China, and has been applied in many aspects, such as irradiation sterilization of food, irradiation disinfection of drugs, etc. The primary standard dose apparatus (calorimeter) of absolute measurement method is indispensable in the measurement of electron beam absorption dose of irradiation processing stage. Therefore, a calorimeter can be designed to measure the absorption dose of electron beam in radiation processing stage. High purity graphite is used as absorber, EPS is used as insulation material, and the structure of each part is designed according to MCNP absorption dose simulation results. The specific dimensions of each part are determined by using ANSYS simulation analysis software. Based on the calorimeter, the resistance resistance of the absorber was measured by using a high precision thermometer bridge of microk250. The results of the measurement of the electron beam absorption dose at 10 MeV nominal energy were obtained. The ratio of the time change of the absorber to the amount of absorbent was obtained. The influence factors of calorimeter accuracy were discussed. The total uncertainty of measuring electron beam absorption dose at energy of 10 MeV was 1.43. The results provide a reference for the measurement of absorbed dose of radiation processing electron beam.

In order to monitor human blood flow rate and flow velocity distribution, a blood flow meter based on multi-electrode electromagnetic measurement has been proposed to measure the flow velocity of the arteriovenous blood by obtaining the induced voltage on the skin surface. The COMSOL 3-D model has been built to simulate the physical structure of the human limb, including the size and relative position of skin, fat, bones, muscles, arteries and veins. Based on the Helmholtz coil and C-type coil, the electromagnetic induction intensity distribution of excitation structures and methods is simulated which is imposed through the body measurement section. The optimal design scheme of the excitation coil is proposed. Different excitation systems of different structures and different incentive methods are compared. The result shows that the two pairs of Helmholtz coils of orthogonal layout with syntroph excitation current has better uniformity, field center symmetry and stronger magnetic field strength, and effectively enhances the value of induction electromotive force, showing it is more suitable for multi-electrode electromagnetic measurement of non-axisymmetric flow. The research result proves that the optimization scheme of excitation system is feasible, and its uniformity plays a positive role in improving the accuracy of velocity reconstruction.

Epilepsy is a chronic disorder of brain dysfunction caused by sudden abnormal discharge of brain neurons. The detection of epileptic seizure can be achieved by the detection and analysis of spike waves. In this paper, a method of detection based on wavelet transform and modulus maximum is proposed. Firstly, it uses the continuous wavelet transform of epileptic EEG signals in a certain scale to divide the frequency bands. Secondly, it applys the modulus algorithm and refining algorithm to detect singular points of EEG signals, which are taken as the suspect points of spike waves. Finally, through screening based on power spectral density analysis and space surface fitting, the final characteristic spike waves are detected to determine whether the epileptic seizure occurres. The verification experimental results indicate the efficiency and reliability of the proposed method with a diagnostic accuracy as high as 92.5%. It provides a valuable reference method for epileptic seizure detection.

Proton density fat fraction (PDFF) is emerging as a useful quantitative magnetic resonance imaging (MRI) biomarker for the diagnosis of hepatic steatosis. In this work, a hybrid method combining complex-based and magnitude-based water-fat separation was proposed to provide accurate measurements of liver PDFF. Clinical experiments were performed to evaluate the feasibility and accuracy of the method proposed. The results demonstrated that PDFF results obtained with the proposed method had excellent correlation and agreement with those calculated from magnetic resonance spectroscopy (MRS).

J wave is an abnormal variation in the electrocardiogram(ECG). It is of great significance to realize J wave automatic classification for the clinical diagnosis of J wave related disease. In this paper, features are extracted based on time-frequency domain and phase space. On the one hand, tunable-Q wavelet transform(TQWT) and high-order cumulant are used to mine the detail characteristics. On the other hand, recurrence plot(RP) is adopted to show the occurrence of recursive points in the cardiac system. After dimensionality reduction, the two types of features, which are fused in parallel, are input to the improved AdaBoost classifier to realize the classification of normal beats, benign J wave beats and malignant J wave beats. The experimental results show that the average accuracy of the classification is about 79%, which is helpful for the diagnosis of benign and malignant J wave.

On the basis of specific irradiation conditions in fusion reactor, the advantages and limitations of the main proposed irradiation facilities including ion irradiation, fission neutron irradiation, spallation neutron irradiation and fusion neutron irradiation were compared and evaluated.Ion accelerators, fission reactors and theory and modeling are expected to continue to play an indispensable role in studying the irradiation effect of fusion materials before the 14MeV neutron source is available. The corresponding relation between irradiation data from real fusion neutron and irradiation data from ion, fission neutron or spallation neutron irradiation has to be verified by irradiation using 14MeV high flux neutron source (e.g. IFMIF). The construction of powerful, economy-efficient fusion materials irradiation facility and the development of multi-scale numerical simulation are the inevitable choices to facilitate the commercialized application of fusion materials.

As a new medical technology, radial shock wave therapy has unique advantages in orthopedics, sports medicine and other medical fields. However, an unified index is needed to evaluate the intensity of clinical treatment of radial shock wave therapy. Based on the principle of radial shock wave generation, the energy flow density was advanced as an index for evaluating the intensity of clinical treatment. High precision underwater dynamic pressure measurement system was established to measure the pressure waveform. Then the laser Doppler vibration measurement was used to measure the vibration wave of the wave source accurately. The correctness of the pressure waveform measured through the experimental system is proved by comparing the two kinds of waveform. Finally, the energy flow density was calculated according to the measured pressure waveform, which is helpful to effectively reduce the blindness because of simultaneous adjustment of multiple parameters, and it was advanced as an index for evaluating the intensity of shock wave, which has positive guiding significance for its clinical application.

Lutetium (Lu) based scintillators such as LSO and LYSO, are widely used in PET detectors due to their high stopping power for 511 keV gamma rays, high light yield and short decay time. The abundance of Lu-176 is 2.6% in natural Lu, and the Lu-176 is a long-lived radioactive isotope with a beta decay and three major simultaneous gamma decays. This phenomenon introduces random events to PET systems that affects the system imaging performance. On the other hand, the advantages of intrinsic radiation of Lu-176 continue to be exploited and utilized. The Lu-176 intrinsic radiation in PET detectors are reviewed, including the characteristic, the adverse effects, and its useful applications.

Multimodal image registration method based on feature matching can't satisfy the demands of pixel level registration precision.This paper proposes a multimodal image registration algorithm considering grayscale and gradient information.The nonparametric registration model based on Markov random field (MRF) makes full use of the image information of multimodal image to measure the similarity which considers the grayscale and the gradient statistical information are considered,and the value space is discretized to improve the convergence speed.The algorithm is validated both qualitatively and quantitatively demonstrating its potentials on three groups of multimodal image registration experiments.The result indicates that the proposed algorithm is superior to the polynomial model registration based on manual selection and the multimodal image registration only with gray information only.At the same time,this algorithm has some applicability for multimodal image registration of large deformation.In terms of spatial accuracy,the average registration error is less than 1 pixel and the maximum registration error is less than 2 pixels.

To validate and determine the contamination source term of fast neutrons produced by the first in-Hospital Neutron Irradiator (IHNI-1) for boron neutron capture therapy (BNCT), a boron-covered 235U fission chamber was designed to be used for the measurement of fast neutron fluence rate. The fluence response of the fission chamber was optimized according to the calculated fluence response curve of the fission chamber covered by the boron shells with different thicknesses with the MCNP code. Consequently, a 35 mm thick B4C shell was used to shield the low energy neutrons. The fast neutron fluence rate of the thermal and epithermal neutron beams was measured with the fission chamber and compared with the simulated values. The results showed that the fast neutron component in the actual neutron beams was more than that in the simulated ones, less than the target value recommended by IAEA.

Ultrasonic fatigue specimen of MB8 wrought magnesium alloy was designed by calculation. In order to reduce the vibration amplitude, heat and pressure of the cooling system and also to protect the phase change, the sample was designed into variable cross-section. Based on the vibration displacement equation, boundary condition and displacement continuity condition, the relationship between the resonant length and other dimensions of the sample was derived. A series of size of ultrasonic fatigue specimens was calculated, which provided reference for the research of ultrasonic fatigue test of magnesium alloy. The results show that the resonance length decreases with the increase of the length of transition section, and the arc transition radius increases with the increase of the length of arc transition section.With the decrease of the length of arc transition section, Cs and M decreases for the specimen without uniform cross section in middle. With the decrease of the length of uniform cross section in the middle, Cs and M decreased for the specimen with uniform cross section in the middle.