Lung nodule volumetry is used for nodule diagnosis, as well as for monitoring tumor response to therapy. Volume measurement precision and accuracy depend on a number of factors, including image-acquisition and reconstruction parameters, nodule characteristics, and the performance of algorithms for nodule segmentation and volume estimation. The purpose of this article is to provide a review of published studies relevant to the computed tomographic (CT) volumetric analysis of lung nodules. A number of underexamined areas of research regarding volumetric accuracy are identified, including the measurement of nonsolid nodules, the effects of pitch and section overlap, and the effect of respiratory motion. The need for public databases of phantom scans, as well as of clinical data, is discussed. The review points to the need for continued research to examine volumetric accuracy as a function of a multitude of interrelated variables involved in the assessment of lung nodules. Understanding and quantifying the sources of volumetric measurement error in the assessment of lung nodules with CT would be a first step toward the development of methods to minimize that error through system improvements and to correctly account for any remaining error.

A number of interrelated factors can affect the precision and accuracy of lung nodule size estimation. To quantify the effect of these factors, we have been conducting phantom CT studies using an anthropomorphic thoracic phantom containing a vasculature insert to which synthetic nodules were inserted or attached. Ten repeat scans were acquired on different multi-detector scanners, using several sets of acquisition and reconstruction protocols and various nodule characteristics (size, shape, density, location). This study design enables both bias and variance analysis for the nodule size estimation task. The resulting database is in the process of becoming publicly available as a resource to facilitate the assessment of lung nodule size estimation methodologies and to enable comparisons between different methods regarding measurement error. This resource complements public databases of clinical data and will contribute towards the development of procedures that will maximize the utility of CT imaging for lung cancer screening and tumor therapy evaluation.

This work is a part of our more general effort to probe the interrelated factors impacting the accuracy and precision of lung nodule measurement tasks. For such a task a low-bias size estimator is needed so that the true effect of factors such as acquisition and reconstruction parameters, nodule characteristics and others can be assessed. Towards this goal, we have developed a matched filter based on an adaptive model of the object acquisition and reconstruction process. Our model derives simulated reconstructed data of nodule objects (templates) which are then matched to computed tomography data produced from imaging the actual nodule in a phantom study using corresponding imaging parameters. This approach incorporates the properties of the imaging system and their effect on the discrete 3-D representation of the object of interest. Using a sum of absolute differences cost function, the derived matched filter demonstrated low bias and variance in the volume estimation of spherical synthetic nodules ranging in density from -630 to +100 HU and in size from 5 to 10 mm. This work could potentially lead to better understanding of sources of error in the task of lung nodule size measurements and may lead to new techniques to account for those errors.

The article concludes with a brief discussion of moving from the assessment of measurement variation to the steps necessary to establish the efficacy of a metric as a biomarker for response.

A multiple circular path convolution neural network (MCPCNN) architecture specifically designed for the analysis of tumor and tumor-like structures has been constructed. We first divided each suspected tumor area into sectors and computed the defined mass features for each sector independently. These sector features were used on the input layer and were coordinated by convolution kernels of different sizes that propagated signals to the second layer in the neural network system. The convolution kernels were trained, as required, by presenting the training cases to the neural network. In this study, randomly selected mammograms were processed by a dual morphological enhancement technique. Radiodense areas were isolated and were delineated using a region growing algorithm. The boundary of each region of interest was then divided into 36 sectors using 36 equi-angular dividers radiated from the center of the region. A total of 144 Breast Imaging-Reporting and Data System-based features (i.e., four features per sector for 36 sectors) were computed as input values for the evaluation of this newly invented neural network system. The overall performance was 0.78-0.80 for the areas (Az) under the receiver operating characteristic curves using the conventional feed-forward neural network in the detection of mammographic masses. The performance was markedly improved with Az values ranging from 0.84 to 0.89 using the MCPCNN. This paper does not intend to claim the best mass detection system. Instead it reports a potentially better neural network structure for analyzing a set of the mass features defined by an investigator.

No abstract

No abstract

Our purpose in this work was to develop an automatic boundary detection method for mammographic masses and to rigorously test this method via statistical analysis. The segmentation method utilized a steepest change analysis technique for determining the mass boundaries based on a composed probability density cost function. Previous investigators have shown that this function can be utilized to determine the border of the mass body. We have further analyzed this method and have discovered that the steepest changes in this function can produce mass delineations that include extended projections. The method was tested on 124 digitized mammograms selected from the University of South Florida's Digital Database for Screening Mammography (DDSM). The segmentation results were validated using overlap, accuracy, sensitivity, and specificity statistics, where the gold standards were manual traces provided by two expert radiologists. We have concluded that the best intensity threshold corresponds to a particular steepest change location within the composed probability density function. We also found that our results are more closely correlated with one expert than with the second expert. These findings were verified via Analysis of Variance (ANOVA) testing. The ANOVA tests obtained p-values ranging from 1.03 x 10(-2)-7.51 x 10(-17) for the single observer studies and 2.03 x 10(-2)-9.43 x 10(-4) for the two observer studies. Results were categorized using three significance levels, i.e., p<0.001 (extremely significant), p <0.01 (very significant), and p <0.05 (significant), respectively.