Performance comparison of breast imaging modalities using a 4AFC human observer study

Elangovan, Premkumar; Rashidnasab, Alaleh; Mackenzie, Alistair; Dance, David R.; Young, Kenneth C.; Bosmans, Hilde; Segars, W. Paul; Wells, Kevin

doi:10.1117/12.2081878

Cited by 10 publications

(10 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The only exception was for 4mm objects and DBT, where the observers found irregular masses easier to detect than spheres resulting in low threshold contrast for masses than spheres (p<0.0001). This result is consistent with our previous study where 3mm spheres and masses were used in a 4AFC study (Elangovan et al 2015). This may be because the effect of background texture on object appearance is small for very small objects, particularly in DBT, and as a result irregular objects with irregular edges would have been more easily detectable than regular objects such as spheres.…”

Section: Figuresupporting

confidence: 92%

Lesion detectability in 2D-mammography and digital breast tomosynthesis using different targets and observers

et al. 2018

Self Cite

View full text Add to dashboard Cite

This work investigates the detection performance of specialist and non-specialist observers for different targets in 2D-mammography and digital breast tomosynthesis (DBT) using the OPTIMAM virtual clinical trials (VCT) Toolbox and a 4-alternative forced choice (4AFC) assessment paradigm. Using 2D-mammography and DBT images of virtual breast phantoms, we compare the detection limits of simple uniform spherical targets and irregular solid masses. Target diameters of 4 mm and 6 mm have been chosen to represent target sizes close to the minimum detectable size found in breast screening, across a range of controlled contrast levels. The images were viewed by a set of specialist observers (five medical physicists and six experienced clinical readers) and five non-specialists. Combined results from both observer groups indicate that DBT has a significantly lower detectable threshold contrast than 2D-mammography for small masses (4 mm: 2.1% [DBT] versus 6.9% [2D]; 6 mm: 0.7% [DBT] versus 3.9% [2D]) and spheres (4 mm: 2.9% [DBT] versus 5.3% [2D]; 6 mm: 0.3% [DBT] versus 2.2% [2D]) (p < 0.0001). Both observer groups found spheres significantly easier to detect than irregular solid masses for both sizes and modalities (p < 0.0001) (except 4 mm DBT). The detection performances of specialist and non-specialist observers were generally found to be comparable, where each group marginally outperformed the other in particular detection tasks. Within the specialist group, the clinical readers performed better than the medical physicists with irregular masses (p < 0.0001). The results indicate that using spherical targets in such studies may produce over-optimistic detection thresholds compared to more complex masses, and that the superiority of DBT for detecting masses over 2D-mammography has been quantified. The results also suggest specialist observers may be supplemented by non-specialist observers (with training) in some types of 4AFC studies.

show abstract

Section: Figuresupporting

confidence: 92%

Lesion detectability in 2D-mammography and digital breast tomosynthesis using different targets and observers

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…Other authors have taken different approaches to the challenge of simulating realistic breast textures. Elangovan et al 53 have recently demonstrated how breast structure extracted from reconstructed DBT planes can be combined to produce realistic breast tissue background for observer studies in mammography and DBT. Such an approach is attractive since it takes advantage of the high resolution of a digital flat panel detector combined with the quasi-3D resolution of tomosynthesis.…”

Section: Discussionmentioning

confidence: 99%

Population of 224 realistic human subject-based computational breast phantoms

et al. 2015

View full text Add to dashboard Cite

Purpose: To create a database of highly realistic and anatomically variable 3D virtual breast phantoms based on dedicated breast computed tomography (bCT) data. Methods: A tissue classification and segmentation algorithm was used to create realistic and detailed 3D computational breast phantoms based on 230+ dedicated bCT datasets from normal human subjects. The breast volume was identified using a coarse three-class fuzzy C-means segmentation algorithm which accounted for and removed motion blur at the breast periphery. Noise in the bCT data was reduced through application of a postreconstruction 3D bilateral filter. A 3D adipose nonuniformity (bias field) correction was then applied followed by glandular segmentation using a 3D bias-corrected fuzzy C-means algorithm. Multiple tissue classes were defined including skin, adipose, and several fractional glandular densities. Following segmentation, a skin mask was produced which preserved the interdigitated skin, adipose, and glandular boundaries of the skin interior. Finally, surface modeling was used to produce digital phantoms with methods complementary to the XCAT suite of digital human phantoms. Results: After rejecting some datasets due to artifacts, 224 virtual breast phantoms were created which emulate the complex breast parenchyma of actual human subjects. The volume breast density (with skin) ranged from 5.5% to 66.3% with a mean value of 25.3% ± 13.2%. Breast volumes ranged from 25.0 to 2099.6 ml with a mean value of 716.3 ± 386.5 ml. Three breast phantoms were selected for imaging with digital compression (using finite element modeling) and simple ray-tracing, and the results show promise in their potential to produce realistic simulated mammograms. Conclusions: This work provides a new population of 224 breast phantoms based on in vivo bCT data for imaging research. Compared to previous studies based on only a few prototype cases, this dataset provides a rich source of new cases spanning a wide range of breast types, volumes, densities, and parenchymal patterns.

show abstract

“…Seven human observers performed the 12 detection tasks with different background noise structures ( Table 2). For each detection task, the training session consisted of 30 trials, while the testing session consisted of 100 trials [2], [27]. In each trial, four images were displayed on a Nio 3MP LED monitor (Barco, Kortrijk, Belgium) with 2048×1536 pixel resolution and the human observers were asked to choose the signal-present image among one signal-present and three signal-absent images.…”

Section: The Human Observer Studymentioning

confidence: 99%

A Convolutional Neural Network-Based Anthropomorphic Model Observer for Signal Detection in Breast CT Images Without Human-Labeled Data

2020

View full text Add to dashboard Cite

Various imaging parameters in X-ray computed tomography (CT) should be examined and optimized by task-based assessment of human observer performance. Recently, convolutional neural networks (CNNs) have been introduced as anthropomorphic model observers. However, when humanlabeled data are not available or limited, CNNs with an existing training strategy do not produce good performance agreement with human observers. The purpose of this study is to propose new training strategies for a CNN-based anthropomorphic model observer without human-labeled data for signal-knownexactly and background-known-statistically detection tasks. We acquired cone-beam CT projection data of breast background volume and reconstructed the projection data using the Feldkamp-Davis-Kress algorithm with 12 different imaging conditions including viewing image planes. Training data for the CNN were labeled utilizing conventional model observers. We employed an early stopping rule to reflect internal noise during the CNN training. To examine the CNN performance, we used three different training-testing schemes. The performance agreement between the human and model observers was measured via a Bland-Altman plot, the root-mean-squared error (RMSE), and the Pearson's correlation coefficient (r) of their proportion correct values. Throughout the three different training-testing schemes, CNNs with the proposed training strategies yielded narrower limits of agreements (with a bias lower than 0.03) and higher scores in both RMSE and r than the conventional model observers. This indicates the proposed training strategies enable the CNN-based anthropomorphic model observer to have good performance agreement with human observers and generalize better to different imaging conditions than conventional anthropomorphic model observers. INDEX TERMS Anthropomorphic model observer, breast cone-beam CT, convolutional neural network, data labeling, early stopping, signal detection.

show abstract

Performance comparison of breast imaging modalities using a 4AFC human observer study

Cited by 10 publications

References 9 publications

Lesion detectability in 2D-mammography and digital breast tomosynthesis using different targets and observers

Lesion detectability in 2D-mammography and digital breast tomosynthesis using different targets and observers

Population of 224 realistic human subject-based computational breast phantoms

A Convolutional Neural Network-Based Anthropomorphic Model Observer for Signal Detection in Breast CT Images Without Human-Labeled Data

Contact Info

Product

Resources

About