Multicriteria maximum likelihood neural network approach to positron emission tomography

The emerging technology of positron emission image reconstruction is introduced in this paper as a multicriteria optimization problem. We show how selected families of objective functions may be used to reconstruct positron emission images. We develop a novel neural network approach to positron emission imaging problems. We also studied the most frequently used image reconstruction methods, namely, maximum likelihood under the framework of single performance criterion optimization. Finally, we introduced some of the results obtained by various reconstruction algorithms using computer‐generated noisy projection data from a chest phantom and real positron emission tomography (PET) scanner data. Comparison of the reconstructed images indicated that the multicriteria optimization method gave the best in error, smoothness (suppression of noise), gray value resolution, and ghost‐free images. © 2001 John Wiley & Sons, Inc. Int J Imaging Syst Technol 11, 361–364, 2000     

A cross-entropy approach to the single row facility layout problem

The single row facility layout problem is to arrange a given number of facilities along a straight line so as to minimise the total cost associated with the interactions between the facilities. In this paper, a metaheuristic algorithm based on the cross-entropy method, incorporating a local search procedure and symmetry-breaking techniques, is developed to solve this problem. The proposed algorithm has been tested on some widely used benchmark instances. The computational results show that the proposed algorithm has found the optimal or the best solutions known so far for the instances of size with up to 100 facilities and is competitive with some existing algorithms.     

SQUID tomographic neuromagnetic imaging

The human brain emits a measurable magnetic field in response to stimulation. The aim of neuromagnetic imaging is to produce maps of the underlying neuronal activity from measurements of the emitted magnetic field. Modeling neuronal activity by dipolar current sources, we have previously reported simulation studies to reconstruct planar source distributions using the algebraic reconstruction technique, with the constraint that the sources could only have two orientations. We now present a general solution for two-dimensionally distributed sources, which is based on measuring both the normal and tangential components of the magnetic field. Also, using a single-channel SQUID neuromagnetometer, we present results of a tomographic imaging experiment conducted with wires carrying currents to demonstrate the potential of neuromagnetic imaging in reconstructing extended cortical sources.     

A time-domain approach to reactive current minimization innonsinusoidal situations

For the optional LC compensator design, the RMS (root mean square) values of the sources voltage harmonics and the load susceptances, have to be measured. It is shown that the optimal capacitance of such an LC compensator can be determined in the time domain without this information. This can be done by an iterative procedure which requires that the scalar product of the load current and the derivative of the capacitor voltage and their RMS values be measured     

模拟生物样品的反射式超声CT成像

本文简要介绍了反射式超声ＣＴ成像方法的原理，介绍了远场条件下的ＵＲＣＴ成像算法。对模拟生物样品进行了ＵＲＣＴ成像实验，并对多辐二维ＵＲＣＴ图像，得到了ＵＲＣＴ图像的三维显示。     

Fuzzy-rule-based image reconstruction for positron emission tomography

Positron emission tomography (PET) and single-photon emission computed tomography have revolutionized the field of medicine and biology. Penalized iterative algorithms based on maximum a posteriori (MAP) estimation eliminate noisy artifacts by utilizing available prior information in the reconstruction process but often result in a blurring effect. MAP-based algorithms fail to determine the density class in the reconstructed image and hence penalize the pixels irrespective of the density class. Reconstruction with better edge information is often difficult because prior knowledge is not taken into account. The recently introduced median-root-prior (MRP)-based algorithm preserves the edges, but a steplike streaking effect is observed in the reconstructed image, which is undesirable. A fuzzy approach is proposed for modeling the nature of interpixel interaction in order to build an artifact-free edge-preserving reconstruction. The proposed algorithm consists of two elementary steps: (1) edge detection, in which fuzzy-rule-based derivatives are used for the detection of edges in the nearest neighborhood window (which is equivalent to recognizing nearby density classes), and (2) fuzzy smoothing, in which penalization is performed only for those pixels for which no edge is detected in the nearest neighborhood. Both of these operations are carried out iteratively until the image converges. Analysis shows that the proposed fuzzy-rule-based reconstruction algorithm is capable of producing qualitatively better reconstructed images than those reconstructed by MAP and MR P algorithms. The reconstructed images a resharper, with small features being better resolved owing to the nature of the fuzzy potential function.     

Resolution enhancement in computerized tomographic imaging

We consider the problem of reconstructing an object function f(r) from finitely many linear functional values. In our main application, the function f(r) is a tomographic image, and the data are integrals of f(r) along thin strips. Because the data are limited, resolution can be enhanced through the inclusion of prior knowledge. One way to do that, a generalization of the prior discrete Fourier transform (PDFT) method, was suggested in 1982 [SIAM J. Appl. Math.42,933 (1982)] but was found to be difficult to implement for the tomography problem, and that application was not pursued. Recent advances in approximating the PDFT make it possible to achieve the desired resolution enhancement in an easily implemented procedure.     

A method to modify coordinates of detectors in positron emission tomography systems

A new method to modify coordinates of detectors in any positron emission tomography (PET) system using a radioactive point source is developed. This method is based on selecting a centered detector in each detector block of PET and measuring coincidence counts between the two centered detectors in opposite detector blocks to find the coordinates of their LOR (Line of Response). Due to slight misalignment of detector positions, measured LORs may not intersect at a single point. Based on the proposed method, the coordinates of detectors can be measured with very high accuracy and the coordinate of the center of the gantry (which is normally the same as the center of field of view) can be defined correctly. The results of the application of our method to a small animal PET system (FinePET), which was recently developed at Tohoku University, Japan, are shown here. The method is expected to contribute to the design and development of PET systems which can realize a very high spatial resolution of less than 1 mm FWHM.     

A compact planar positron imaging system

A planar imaging system for positron-emitting radiotracers was developed and its physical performance was evaluated. The new device consists of two opposing planar detectors, each having 4 (columns) $\text{×}\text{6}$ (rows) detector units, and each unit composed of 10×10 arrays of $\text{2×2×20}\text{mm}{}^3$ pillars of Bi₄Ge₃O₁₂ (BGO) scintillators and a metal-packaged position-sensitive photomultiplier tube. The system is very compact and has a simple structure that allows versatile positioning of detectors in the horizontal or vertical mode. Focal plane images are constructed from coincidence data collected by opposing planar detectors. The system provides good quality images even in a short period of time or with low tracer activity. Spatial resolutions of 1.6–$\text{2.1}\text{mm}$ FWHM were achieved for the entire field of view. The sensitivity was $\text{107}\text{cps}\text{/}\text{kBq}\text{/}\text{ml}$ measured in a uniform flat phantom. In this study, the capability of the new system was demonstrated by the imaging of radiotracer distribution in rats and plants.     

Neural network-based image reconstruction for positron emission tomography

Positron emission tomography (PET) is one of the key molecular imaging modalities in medicine and biology. Penalized iterative image reconstruction algorithms frequently used in PET are based on maximum-likelihood (ML) and maximum a posterior (MAP) estimation techniques. The ML algorithm produces noisy artifacts whereas the MAP algorithm eliminates noisy artifacts by utilizing availableprior information in the reconstruction process. The MAP-based algorithms fail to determine the density class in the reconstructed image and hence penalize the pixels irrespective of the density class and irrespective of the strength of interaction between the nearest neighbors. A Hebbian neural learning scheme is proposed to model the nature of interpixel interaction to reconstruct artifact-free edge preserving reconstruction. A key motivation of the proposed approach is to avoid oversmoothing across edges that is often the case with MAP algorithms. It is assumed that local correlation plays a significant role in PET image reconstruction, and proper modeling of correlation weight (which defines the strength of interpixel interaction) is essential to generate artifact-free reconstruction. The Hebbian learning-based approach modifies the interaction weight by adding a small correction that is proportional to the product of the input signal (neighborhood pixels) and output signal. Quantitative analysis shows that the Hebbian learning-based adaptive weight adjustment approach is capable of producing better reconstructed images compared with those reconstructed by conventional ML and MAP-based algorithms in PET image reconstruction.     

Developments in high-resolution positron emission tomography at MGH

The development of high resolution PET systems is important for the wider application of this techniques. The resolution of PET is limited by a number of physical factors such as positron range, small angle deviation, and sampling frequency. The design of the detector array and its sensitivity remain critical factors; designs incorporating analog coding have proven to be useful. PCR-I, a single plane PET system, has demonstrated the concept and has produced useful images in animal studies. PCR-II will extend the concept to a two dimensional detector array resulting in a system with high resolution and high sensitivity.     

Air-coupled ultrasonic tomographic imaging of high-temperature flames

This paper illustrates the use of air-coupled ultrasonic tomography for the measurement of a highs temperature flame from a natural gas burner, using capacitive ultrasonic transducers in through transmission. This uses a transducer pair, which is scanned in two-dimensional sections at several angles to the jet axis. Travel-time data then is recorded along various paths in counter-propagating directions. By processing the data obtained from propagation times, images have been formed of variations in temperature within the flame, using the tomographic reconstruction approach.     

A calibration approach to acoustic emission energy measurement

An approximate method for the quantitative characterization of acoustic emission sources whose signals are recorded with traditional instrumentation has been developed to evaluate the emission from a Fe-Ni alloy undergoing martensitic transformations. The calibration principle is based upon using pseudo sources of known elastic strain energy to produce a calibration curve relating a parameter of the acoustic emission signal measured (envelope strength) to the strain energy of the source. The curve can then be used to determine the strain energy of naturally occurring sources during martensitic transformations.     

Projection approach to complexity reduction in tomographic alignment of extremely large telescopes

We describe a complexity reduction approach intended to solve the tomographic alignment problem for the Thirty Meter Telescope by means of its alignment and phasing system (APS) with little loss of information. This approach is computationally efficient enough to perform detailed Monte-Carlo simulations of the APS on a standard PC. We present sample simulations to model error propagation through the system and to build a preliminary APS alignment error budget.     

Numerical model of optical coherence tomographic vibrography imaging to estimate corneal biomechanical properties

Most techniques measuring corneal biomechanics in vivo are biased by side factors. We demonstrate the ability of optical coherence tomographic (OCT) vibrography to determine corneal material parameters, while reducing current prevalent restrictions of other techniques (such as intraocular pressure (IOP) and thickness dependency). Modal analysis was performed in a finite-element (FE) model to study the oscillation response in isolated thin corneal flaps/eye globes and to analyse the dependency of the frequency response function on: corneal elasticity, viscoelasticity, geometry (thickness and curvature), IOP and density. The model was verified experimentally in flaps from three bovine corneas and in two enucleated porcine eyes using sound excitation (100–110 dB) together with a phase-sensitive OCT to measure the frequency response function (range 50–510 Hz). Simulations showed that corneal vibration in flaps is sensitive to both, geometrical and biomechanical parameters, whereas in whole globes it is primarily sensitive to corneal biomechanical parameters only. Calculations based on the natural frequency shift revealed that flaps of the posterior cornea were 0.8 times less stiff than flaps from the anterior cornea and cross-linked corneas were 1.6 times stiffer than virgin corneas. Sensitivity analysis showed that natural vibration frequencies of whole globes were nearly independent from corneal thickness and IOP within the physiological range. OCT vibrography is a promising non-invasive technique to measure corneal elasticity without biases from corneal thickness and IOP.     

A BaF2-MWPC gamma camera for positron emission tomography

The construction and performance of a position sensitive detector for 511 keV annihilation radiation are described. The detector consists of a barium fluoride crystal sandwiched between two wire chambers which are operated with 45 and 50°C TMAE vapour. Energy, time and position resolution are presented. The advantages and limitations of the detector are discussed. We also present some preliminary data on other UV scintillators.     

Sampling-theory approach to eigenwavefronts of imaging systems

We present a direct method based on the sampling theorem for computing eigenwavefronts associated with linear space-invariant imaging systems (including aberrated imaging systems). A potential application of the eigenwavefronts to inverse problems in imaging is discussed. A noise-dependent measure for the information-carrying capacity of an imaging system is also proposed.     

Characterization of a 82Rb generator for positron emission tomography

Characteristics of the developed ⁸²Rb medical generator intended for the production of Rubidium Chloride, ⁸²Rb from Generator radiopharmaceutical for positron emission tomography (PET) are examined. The results of laboratory tests and clinical trials are presented. The influence of thermal sterilization on the ionexchange characteristics of the sorbent based on hydrous tin(IV) oxide is studied. The possibility of radiation “self-sterilization” of the generator column and of the maintenance of aseptic conditions in it throughout the generator operation life is assessed. In the course of operation of one generator, it is possible to obtain up to 19 L of a sterile and apyrogenic solution of the radiopharmaceutical with permissible content of radioactive and stable impurities. The efficiency of the ⁸²Rb elution from the sorbent at different eluent flow rates is determined. The efficiency of using the generator in PET investigations in cardiology and oncology is demonstrated.     

A neural network approach to microwave imaging

We present a neural network approach to microwave imaging for medical diagnosis. The problem is to reconstruct the complex permittivity of the biological tissues illuminated by the transverse magnetic (TM) incident waves. In order to avoid the inherent ill‐posedness of the inverse scattering problem, we introduce a stochastic process based on Markov random field and a priori knowledge. A coupled gradient neural network is proposed to deal with the mixed‐variable problem because the reconstructed dielectric permittivities are continuous complex variables and the line processes, which can preserve the edges of the reconstructed image, are binary variables. We report the numerical results of a simple human forearm model. We also point out the advantages and the limitations of this method. © 2001 John Wiley & Sons, Inc. Int J Imaging Syst Technol 11, 159–163, 2000