Biomedical subjects

J M Boone

Publications and source records attributed to J M Boone.

At least 19 recordsLinked to original sources

PubMed1992

Recognition of chest radiograph orientation for picture archiving and communications systems display using neural networks.

A neural network classification scheme was developed that enables a picture archiving and communications system workstation to determine the correct orientation of posteroanterior or anteroposterior chest images. This technique permits thoracic images to be displayed conventionally when called up on the workstation, and therefore reduces the need for reorientation of the image by the observer. Feature data were extracted from 1,000 digitized chest radiographs and used to train a two-layer neural network designed to classify the image into one of the eight possible orientations for a posteroanterior chest image. Once trained, the neural network identified the correct image orientation in 888 of 1,000 images that had not previously been seen by the neural network. Of the 112 images that were incorrectly classified, 106 were mirror images of the correct orientation, whereas only 6 actually had the caudal-cranial axis aligned incorrectly. The causes for misalignment are discussed.

Humans↗

PubMed1992

Binary screen detector system for single-pulse dual-energy radiography.

Dual-energy radiographic acquisition performed with a single pulse of x rays has been limited to use of stacked detectors such as photo-stimulable phosphor plates. In this study, a binary screen system is introduced that enables single-pulse dual-energy acquisition with nonstacked detectors such as charge-coupled devices (CCDs). Two x-ray phosphors with different K edges, designed to emit light of different wavelengths, were bound together in a single screen. Two CCD cameras, each sensitive to the respective wavelengths emitted by each phosphor, are proposed to be coupled to the binary screen. Optical isolation and detection of the optical emissions from the screen would enable simultaneous acquisition of both low- and high-energy images, which can then be used in dual-energy subtraction. Computer simulation results and experimentally acquired images suggest that the binary screen approach may be a viable alternative to stacked detector technology for dual-energy radiographic imaging.

Computer Simulation↗

PubMed1991

Color mammography. Image generation and receiver operating characteristic evaluation.

Color mammography is a technique whereby dual-energy mammographic image data are used to calculate a calcium image; the calcium image is colorized and overlaid onto the conventional (lower energy) gray scale mammogram for radiologist viewing. This technique is presented as a practical way to use the increased calcium sensitivity of dual-energy mammography without requiring an increase in the number of images that the radiologist must read, and without subjecting the radiologist to the unfamiliar appearance of the dual-energy subtracted images. Using straightforward imaging theory, the acquisition techniques for both conventional and dual-energy mammography were optimized, and the optimal technique factors were used to generate a series of computer-simulated mammographic images that were used in a receiver operating characteristic (ROC) comparative study. Ideal observer ROC experiments indicate that (dual-energy) calcium images yield consistently higher sensitivity and specificity to the presence of calcifications, regardless of the amount of tissue "clutter," whereas conventional mammography results show degraded detectability performance as tissue contrast increases. Using human observers viewing simulated images, color mammography delivered greater calcification detectability than conventional mammography.

Breast Diseases↗

PubMed1991

Radiation exposure to angiographers under different fluoroscopic imaging conditions.

Radiation levels near an imaging chain commonly used in angiography were measured with both a 100- and a 200-mm-thick scatter phantom. The scatter was measured in lines parallel in space to the central ray of the x-ray beam, at lateral distances of 300, 500, and 800 mm. The effects of fluoroscopic kilovoltage and image intensifier magnification mode were also measured. The results indicate that the highest scattered radiation levels occur near the surface of the patient where the x-ray beam enters. Exposure rates were measured in both anteroposterior (AP) and posteroanterior (PA) geometries on a U-arm system. In PA geometry, the highest radiation levels occur below the angiographer's waist, an area well protected by the lead apron. The AP geometry increases the exposure rate to the neck, head, and upper extremities, areas where apron shielding is less effective.

Angiography↗

PubMed1990

Neural networks in radiologic diagnosis. I. Introduction and illustration.

Artificial neural networks (NNs) process information in a manner similar to the way the human brain is thought to process information. Neural networks have potential application in radiology as an artificial intelligence technique that can provide computer-aided diagnostic assistance for the practicing radiologist. The basic characteristics of NNs and the manner in which information propagates through an NN are discussed in nontechnical language, to assist the diagnostic radiologist in understanding the basic principles of neurocomputing. Computer-aided diagnosis selection in pediatric chest radiography using NNs is discussed in a companion article.

Artificial Intelligence↗

PubMed1990

Neural networks in radiologic diagnosis. II. Interpretation of neonatal chest radiographs.

A neural network (NN) system was trained to choose one or more diagnoses from a list of 12 possible diagnoses, based on 21 radiographic observations made on each of a series of neonatal chest radiographs. Initially, an experienced pediatric radiologist provided both the radiographic observations and ranked differential diagnoses for each of 77 neonatal chest radiographs in the preliminary phase used to train the NN. Subsequently, two pediatric radiologists (one of whom provided the initial training-phase data) independently read a series of 103 neonatal chest radiographs (different from the training set) and compiled a list of radiographic findings and differential diagnoses for each radiograph. The trained NN was then asked to provide a list of differential diagnoses for each case from the radiologists' lists of findings. Agreement between the network and each radiologist independently was greater than between the two radiologists. Both the positive and negative agreement between the network and either radiologist was greater than the inter-radiologist agreements for most of the diagnostic endpoints.

Artificial Intelligence↗

PubMed1988

Scattered energy deposition under shielding.

Monte Carlo methods are used to generate line spread functions describing dose distributions at a variety of depths within a homogeneous water phantom. The line spread function data are convolved with a step function that represents the edge of a primary radiation field. The dosimetric information beyond the edge of the field is reported in the form of tissue-air ratios for three different beam spectra in the diagnostic energy range.

Models, Structural↗

PubMedDate not supplied

Equivalent spectra as a measure of beam quality.

The concept of the equivalent spectrum (Seq) is introduced, where Seq is defined as an idealized energy spectrum which results in identical attenuation properties as the actual spectrum of a given x-ray tube. Seq is described by two parameters, the equivalent aluminum filtration Aleq and the equivalent kilovoltage kVeq, which are evaluated iteratively using aluminum attenuation data. Published spectral data are used to demonstrate the merits of the equivalent spectrum technique, and experimental results are also presented. The technique is shown to be sensitive to changes in total filtration and, indirectly, to the waveform properties of the x-ray system. The use of the Seq may result in a better description of spectral properties for reporting purposes, and aid in standardizing radiographic techniques within an institution.

Computer Simulation↗

PubMedDate not supplied

Characterization of the point spread function and modulation transfer function of scattered radiation using a digital imaging system.

A digital radiographic system was used to measure the distribution of scattered x radiation from uniform slabs of Lucite at various thicknesses. Using collimation and air gap techniques, [primary + scatter] images and primary images were digitally acquired, and subtracted to obtain scatter images. The scatter distributions measured using small circular apertures were computer fit to an analytical function, representing the circular aperture function convolved with a modified Gaussian point spread function (PSF). On the basis of goodness of fit criterion, the proposed Gaussian function is a very good model for the scatter PSF. The measured scatter PSF's are reported for various Lucite thicknesses. Using the PSF's, the modulation transfer functions are calculated, and this spatial frequency information may have value in analytical scatter removal techniques, grid design, and air gap optimization.

Computers↗

PubMedDate not supplied

Scatter correction algorithm for digitally acquired radiographs: theory and results.

A scatter correction algorithm for digitally acquired radiographs (SCADAR) is presented. SCADAR requires the acquisition of two digital images, taken at different object-to-detector distances. These two images are digitally magnification compensated, and subtracted. The primary component in the resulting difference image delta is mathematically eliminated, and hence the delta image is used as a measure of the local contribution of scattered radiation. A gray scale transformation is used to transform the delta image to a scattered component image, which is then smoothed by Fourier filtering, using a matched filter to increase the signal-to-noise ratio. The smoothed scatter image is then subtracted from its corresponding original image, resulting in the corrected SCADAR image. Implementation of SCADAR can result in a large increase in image contrast, and a significant reduction in shading due to scattered radiation effects. The mathematical derivation of the algorithm is developed, and experimental verification is given for some of the principles used. Using experimental images acquired from scattering phantoms, the results of SCADAR on various image parameters such as contrast and detail signal-to-noise ratio are discussed.

Biometry↗

PubMedDate not supplied

Monte Carlo simulation of the scattered radiation distribution in diagnostic radiology.

Monte Carlo techniques were employed to evaluate the point spread function (PSF) of scattered radiation in diagnostic radiology. The Monte Carlo procedure is described and shown to compare well with Monte Carlo scatter analysis of other authors. The intensity and distribution of the PSF are described independently. The effects of object thickness, air gap, and beam spectra are examined. An analytic derivation of the scatter PSF is presented in a companion article, and the Monte Carlo results discussed herein are used for comparison.

Biophysical Phenomena↗

PubMedDate not supplied

An analytical model of the scattered radiation distribution in diagnostic radiology.

A simple scatter model is used to analytically derive the point spread function (PSF) for scattered radiation in diagnostic radiology. The resulting equation is a function of four physical parameters; object thickness, object-to-detector distance (air gap), and the linear attenuation coefficients for both primary and scatter radiation. Though the model is based upon single scattering, it is shown that by reducing the scatter attenuation coefficient the analytic model compares well to the multiple scattering PSF determined using Monte Carlo analysis.

Biophysical Phenomena↗

PubMedDate not supplied

X-ray scatter removal by deconvolution.

The distribution of scattered x rays detected in a two-dimensional projection radiograph at diagnostic x-ray energies is measured as a function of field size and object thickness at a fixed x-ray potential and air gap. An image intensifier-TV based imaging system is used for image acquisition, manipulation, and analysis. A scatter point spread function (PSF) with an assumed linear, spatially invariant response is modeled as a modified Gaussian distribution, and is characterized by two parameters describing the width of the distribution and the fraction of scattered events detected. The PSF parameters are determined from analysis of images obtained with radio-opaque lead disks centrally placed on the source side of a homogeneous phantom. Analytical methods are used to convert the PSF into the frequency domain. Numerical inversion provides an inverse filter that operates on frequency transformed, scatter degraded images. Resultant inverse transformed images demonstrate the nonarbitrary removal of scatter, increased radiographic contrast, and improved quantitative accuracy. The use of the deconvolution method appears to be clinically applicable to a variety of digital projection images.

Computer Simulation↗

PubMedDate not supplied

The three parameter equivalent spectra as an index of beam quality.

A parametric spectral model based on the work of Birch and Marshall is used to characterize the x-ray spectra of a specific x-ray system. Using least-squares comparison between measured and calculated attenuation data, an equivalent spectrum (EQSPEC) is iteratively found which very closely matches the measured attenuation characteristics of the x-ray system. The resulting parametric spectrum is a function of the anode angle (theta), equivalent kilovoltage (kVeq), and the equivalent aluminum filtration (Aleq), and these three parameters can serve as very concise yet very accurate indices of beam quality. The utility of the EQSPEC for characterization and reporting of x-ray spectra (and thus beam quality) may have numerous applications in diagnostic imaging procedures where spectral quality is an important consideration.

Algorithms↗

PubMedDate not supplied

X-ray spectral reconstruction from attenuation data using neural networks.

An artificial neural network using input data derived from attenuation measurements was trained to generate spectral profiles (relative number of photons versus energy). Once the relative spectral distribution is reconstructed, absolute spectra (number of photons per unit exposure spectral distribution is reconstructed, absolute spectra (number of photons per unit exposure versus energy) can be calculated. A neural network was trained on spectra generated mathematically using the Birch-Marshall model, combined with attenuation data, calculated from the spectra by numerical integration. Whereas attenuation data can be calculated in a straightforward manner from the x-ray spectra, the reverse is not true. Several neural networks were successfully taught to reconstruct the spectra, given the attenuation data. The networks were tested using kV/inherent filtration combinations that were not in the training set, and the performance of the reconstruction was excellent. Noise in the attenuation data was simulated to test the effects of noise propagation in the reconstruction. The effects of network architecture and data averaging on noise propagation were investigated. Experimentally determined spectral data complied by Fewell were also used to train a neural network, and the results of the reconstruction were also found to be excellent.

Artificial Intelligence↗

PubMedDate not supplied

Neural networks in radiology: an introduction and evaluation in a signal detection task.

Neural networks are a computer architecture, implementable in software or hardware, that allow an entirely new approach to the computerized perception of data. These so-called connectionist models are inspired by what is known about the architecture of biological neurons, in which the "intelligence" or processing capability of the network is a result of the interconnection strengths between large arrays of nonlinear processing nodes. Neural networks are described and then are used to analyze the common radiological problem of pattern recognition on a noisy background. Classical signal detection theory is used to compare network performance against that of human observers, using computer-generated sets of very simple "nodules." The neural network performed with better accuracy, relative to human observer performance, in the detection of this elementary test object. Although these results may not scale up with more complex images, the favorable performance of neural networks at this level suggests that further investigation is warranted.

Artificial Intelligence↗

PubMedDate not supplied

Dual-energy mammography: a detector analysis.

Dual-energy mammography acquisition scenarios employing single-shot techniques are examined using computer simulation. A figure of merit of the signal-to-noise ratio squared over the glandular dose was chosen for the optimization task due to its exposure independence. Doses were evaluated using Monte Carlo techniques. The effects of kilovoltage, prepatient filtration, front detector thickness, mid-detector filtration thickness and composition were studied. Of the six detector pairs studied (Y2O2S/Gd2O2S, SrFBr/BaFBr, Y2O2S/LaOBr, Y2O2S/CaWO4, Y2O2S/YTaO4, and Y2O2S/LuTaO4), Y2O2S/Gd2O2S and SrFBr/BaFBr were found to be the best combinations. The effects of scatter and signal quantization were also examined. An alternative display technique whereby the tissue-subtracted (i.e., calcium) image is colorized and overlaid onto the conventional mammogram is introduced.

Color↗

PubMedDate not supplied

Analysis and correction of imperfections in the image intensifier-TV-digitizer imaging chain.

Image intensifier-television-video digitizer (IITVD) systems are commonly used for digital planar image acquisition in radiology. However, the well-known distortions inherent in these systems limit their utility in research and in some clinical applications where quantitatively correct images are required. Software correction techniques have been implemented which restore both the spatial and grey scale quantitative integrity, allowing IITVD systems to be used as analytical research tools. Previously reported and novel correction techniques were used to reduce veiling glare, pincushion distortion, dc bias, and residual shading effects. The results indicate that excellent quantitative integrity can be achieved when these straightforward artifact reduction techniques are employed.

Analog-Digital Conversion↗