Matthew A. Kupinski

Computerized lesion detection on breast ultrasound

We investigated the use of a radial gradient index (RGI) filtering technique to automatically detect lesions on breast ultrasound. After initial RGI filtering, a sensitivity of 87% at 0.76 false-positive detections per image was obtained on a database of 400 patients (757 images). Next, lesion candidates were segmented from the background by maximizing an average radial gradient (ARD) index for regions grown from the detected points. At an overlap of 0.4 with a radiologist lesion outline, 75% of the lesions were correctly detected. Subsequently, round robin analysis was used to assess the quality of the classification of lesion candidates into actual lesions and false-positives by a Bayesian neural network. The round robin analysis yielded an Az value of 0.84, and an overall performance by case of 94% sensitivity at 0.48 false-positives per image. Use of computerized analysis of breast sonograms may ultimately facilitate the use of sonography in breast cancer screening programs.

Multiobjective genetic optimization of diagnostic classifiers with implications for generating receiver operating characteristic curves

It is well understood that binary classifiers have two implicit objective functions (sensitivity and specificity) describing their performance. Traditional methods of classifier training attempt to combine these two objective functions (or two analogous class performance measures) into one so that conventional scalar optimization techniques can be utilized. This involves incorporating a priori information into the aggregation method so that the resulting performance of the classifier is satisfactory for the task at hand. The authors have investigated the use of a niched Pareto multiobjective genetic algorithm (GA) for classifier optimization. With niched Pareto GAs, an objective vector is optimized instead of a scalar function, eliminating the need to aggregate classification objective functions. The niched Pareto GA returns a set of optimal solutions that are equivalent in the absence of any information regarding the preferences of the objectives. The a priori knowledge that was used for aggregating the objective functions in conventional classifier training can instead be applied post-optimization to select from one of the series of solutions returned from the multiobjective genetic optimization. The authors have applied this technique to train a linear classifier and an artificial neural network (ANN), using simulated datasets. The performances of the solutions returned from the multiobjective genetic optimization represent a series of optimal (sensitivity, specificity) pairs, which can be thought of as operating points on a receiver operating characteristic (ROC) curve. All possible ROC curves for a given dataset and classifier are less than or equal to the ROC curve generated by the niched Pareto genetic optimization.

Objective Assessment of Image Quality

Small-animal imaging has shown great promise in the areas of oncology,cardiology, molecular biology,drug discovery and development, and genetics [Green, 2001]. The demand for small-animal imaging has increased greatly with the recent advances in biomolecular research. Examples include functional genomics, functional protenomics, and molecular targeting of tumor cells or cells with other abnormalities. SPECT and PET imaging are of particular interest because these systems intrinsically image function instead of anatomy. Traditional thought has precluded SPECT and PET imaging of small animals because of the lack of resolution of these systems.In recent years at CGRI and other research facilities, numerous fast, high-resolution and high sensitivity SPECT imaging systems designed specifically for imaging small animals have been developed. These systems produce three-dimensional images of the distribution of radiotracers within the animal and, because these systems have no moving parts, dynamic studies can be readily performed. However, there are many components of such imaging systems that need to be optimized in order to best perform small-animal imaging studies.

Efficiency of the human observer detecting random signals in random backgrounds

The efficiencies of the human observer and the channelized-Hotelling observer relative to the ideal observer for signal-detection tasks are discussed. Both signal-known-exactly (SKE) tasks and signal-known-statistically (SKS) tasks are considered. Signal location is uncertain for the SKS tasks, and lumpy backgrounds are used for background uncertainty in both cases. Markov chain Monte Carlo methods are employed to determine ideal-observer performance on the detection tasks. Psychophysical studies are conducted to compute human-observer performance on the same tasks. Efficiency is computed as the squared ratio of the detectabilities of the observer of interest to the ideal observer. Human efficiencies are approximately 2.1% and 24%, respectively, for the SKE and SKS tasks. The results imply that human observers are not affected as much as the ideal observer by signal-location uncertainty even though the ideal observer outperforms the human observer for both tasks. Three different simplified pinhole imaging systems are simulated, and the humans and the model observers rank the systems in the same order for both the SKE and the SKS tasks.

Maximum-Likelihood Estimation With a Contracting-Grid Search Algorithm

A fast search algorithm capable of operating in multi-dimensional spaces is introduced. As a sample application, we demonstrate its utility in the 2D and 3D maximum-likelihood position-estimation problem that arises in the processing of PMT signals to derive interaction locations in compact gamma cameras. We demonstrate that the algorithm can be parallelized in pipelines, and thereby efficiently implemented in specialized hardware, such as field-programmable gate arrays (FPGAs). A 2D implementation of the algorithm is achieved in Cell/BE processors, resulting in processing speeds above one million events per second, which is a 20× increase in speed over a conventional desktop machine. Graphics processing units (GPUs) are used for a 3D application of the algorithm, resulting in processing speeds of nearly 250,000 events per second which is a 250× increase in speed over a conventional desktop machine. These implementations indicate the viability of the algorithm for use in real-time imaging applications.

Objective comparison of quantitative imaging modalities without the use of a gold standard

Imaging is often used for the purpose of estimating the value of some parameter of interest. For example, a cardiologist may measure the ejection fraction (EF) of the heart in order to know how much blood is being pumped out of the heart on each stroke. In clinical practice, however, it is difficult to evaluate an estimation method because the gold standard is not known, e.g., a cardiologist does not know the true EF of a patient. Thus, researchers have often evaluated an estimation method by plotting its results against the results of another (more accepted) estimation method, which amounts to using one set of estimates as the pseudogold standard. In this paper, we present a maximum-likelihood approach for evaluating and comparing different estimation methods without the use of a gold standard with specific emphasis on the problem of evaluating EF estimation methods. Results of numerous simulation studies will be presented and indicate that the method can precisely and accurately estimate the parameters of a regression line without a gold standard, i.e., without the x axis.

A prototype instrument for single pinhole small animal adaptive SPECT imaging

The authors have designed and constructed a small-animal adaptive SPECT imaging system as a prototype for quantifying the potential benefit of adaptive SPECT imaging over the traditional fixed geometry approach. The optical design of the system is based on filling the detector with the region of interest for each viewing angle, maximizing the sensitivity, and optimizing the resolution in the projection images. Additional feedback rules for determining the optimal geometry of the system can be easily added to the existing control software. Preliminary data have been taken of a phantom with a small, hot, offset lesion in a flat background in both adaptive and fixed geometry modes. Comparison of the predicted system behavior with the actual system behavior is presented, along with recommendations for system improvements.

The multi-module, multi-resolution system (M3R): A novel small-animal SPECT system

We have designed and built an inexpensive, high-resolution, tomographic imaging system, dubbed the multi-module, multi-resolution system, or M3R. Slots machined into the system shielding allow for the interchange of pinhole plates, enabling the system to operate over a wide range of magnifications and with virtually any desired pinhole configuration. The flexibility of the system allows system optimization for specific imaging tasks and also allows for modifications necessary due to improved detectors, electronics, and knowledge of system construction (e.g., system sensitivity optimization). We provide an overview of M3R, focusing primarily on system design and construction, aperture construction, and calibration methods. Reconstruction algorithms will be described and reconstructed images presented.

Feature selection and classifiers for the computerized detection of mass lesions in digital mammography

We have investigated various methods of feature selection for two different data classifiers used in the computerized detection of mass lesions in digital mammograms. Numerous features were extracted from abnormal and normal breast regions from a database consisting of 210 individual mammograms. A step-wise method, a genetic algorithm and individual feature analysis were employed to select a subset of features to be used with linear discriminants. Similar techniques were also employed for an artificial neural network classifier. In both tests the genetic algorithm was able to either outperform or equal the performance of other methods.

Channelized-ideal observer using Laguerre-Gauss channels in detection tasks involving non-Gaussian distributed lumpy backgrounds and a Gaussian signal

We investigate a channelized-ideal observer (CIO) with Laguerre-Gauss (LG) channels to approximate ideal-observer performance in detection tasks involving non-Gaussian distributed lumpy backgrounds and a Gaussian signal. A Markov-chain Monte Carlo approach is employed to determine the performance of both the ideal observer and the CIO using a large number of LG channels. Our results indicate that the CIO with LG channels can approximate ideal-observer performance within error bars, depending on the imaging system, object, and channel parameters. The CIO also outperforms a channelized-Hotelling observer using the same channels. In addition, an alternative approach for estimating the CIO is investigated. This approach makes use of the characteristic functions of channelized data and employs an approximation method to the area under the receiver operating characteristic curve. The alternative approach provides good estimates of the performance of the CIO with five LG channels. However, for large channel cases, more efficient computational methods need to be developed for the CIO to become useful in practice.