Comparison of breast density assessments between synthesized C-View™ & intelligent 2D™ mammography

Multi-scale cascaded networks for synthesis of mammogram to decrease intensity distortion and increase model-based perceptual similarity

PURPOSE

Synthetic digital mammogram (SDM) is a 2D image generated from digital breast tomosynthesis (DBT) and used as a substitute for a full-field digital mammogram (FFDM) to reduce the radiation dose for breast cancer screening. The previous deep learning-based method used FFDM images as the ground truth, and trained a single neural network to directly generate SDM images with similar appearances (e.g., intensity distribution, textures) to the FFDM images. However, the FFDM image has a different texture pattern from DBT. The difference in texture pattern might make the training of the neural network unstable and result in high-intensity distortion, which makes it hard to decrease intensity distortion and increase perceptual similarity (e.g., generate similar textures) at the same time. Clinically, radiologists want to have a 2D synthesized image that feels like an FFDM image in vision and preserves local structures such as both mass and microcalcifications (MCs) in DBT because radiologists have been trained on reading FFDM images for a long time, while local structures are important for diagnosis. In this study, we proposed to use a deep convolutional neural network to learn the transformation to generate SDM from DBT.

METHOD

To decrease intensity distortion and increase perceptual similarity, a multi-scale cascaded network (MSCN) is proposed to generate low-frequency structures (e.g., intensity distribution) and high-frequency structures (e.g., textures) separately. The MSCN consist of two cascaded sub-networks: the first sub-network is used to predict the low-frequency part of the FFDM image; the second sub-network is used to generate a full SDM image with textures similar to the FFDM image based on the prediction of the first sub-network. The mean-squared error (MSE) objective function is used to train the first sub-network, termed low-frequency network, to generate a low-frequency SDM image. The gradient-guided generative adversarial network's objective function is to train the second sub-network, termed high-frequency network, to generate a full SDM image with textures similar to the FFDM image.

RESULTS

1646 cases with FFDM and DBT were retrospectively collected from the Hologic Selenia system for training and validation dataset, and 145 cases with masses or MC clusters were independently collected from the Hologic Selenia system for testing dataset. For comparison, the baseline network has the same architecture as the high-frequency network and directly generates a full SDM image. Compared to the baseline method, the proposed MSCN improves the peak-to-noise ratio from 25.3 to 27.9 dB and improves the structural similarity from 0.703 to 0.724, and significantly increases the perceptual similarity.

CONCLUSIONS

The proposed method can stabilize the training and generate SDM images with lower intensity distortion and higher perceptual similarity.

Opinions on the Assessment of Breast Density Among Members of the Society of Breast Imaging

OBJECTIVE

Dense breast decreases the sensitivity and specificity of mammography and is associated with an increased risk of breast cancer. We conducted a survey to assess the opinions of Society of Breast Imaging (SBI) members regarding density assessment.

METHODS

An online survey was sent to SBI members twice in September 2020. The survey included active members who were practicing radiologists, residents, and fellows. Mammograms from three patients were presented for density assessment based on routine clinical practice and BI-RADS fourth and fifth editions. Dense breasts were defined as heterogeneously or extremely dense. Frequencies were calculated for each survey response. Pearson's correlation coefficient was used to evaluate the correlation of density assessments by different definitions.

RESULTS

The survey response rate was 12.4% (357/2875). For density assessments, the Pearson correlation coefficients between routine clinical practice and BI-RADS fourth edition were 0.05, 0.43, and 0.12 for patients 1, 2, and 3, respectively; these increased to 0.65, 0.65, and 0.66 between routine clinical practice and BI-RADS fifth edition for patients 1, 2, and 3, respectively. For future density grading, 79.0% (282/357) of respondents thought it should reflect both potential for masking and overall dense tissue for risk assessment. Additionally, 47.1% (168/357) of respondents thought quantitative methods were of use.

CONCLUSION

Density assessment varied based on routine clinical practice and BI-RADS fourth and fifth editions. Most breast radiologists agreed that density assessment should capture both masking and overall density. Moreover, almost half of respondents believed computer or artificial intelligence-assisted quantitative methods may help refine density assessment.