br In our study four features were

In our study, four features were identified as most significant predictors of the subtypes: roundness, concavity, gray mean, and correlation. Roundness depicts a lesion's shape, whereas concavity reflects the irregularity of a lesion boundary. Gray mean simply represents the brightness of a lesion. Correlation represents the smoothing gradient of patterns, and a larger value indicates a smoother pattern. Based on these interpreta-tions, our results indicate that triple-negative breast cancers may be more round, regular, and brighter than other subtypes. Similarly, our results imply that luminal lesions may be more smoothing than HER2-enriched and triple-negative breast cancers. These findings, however, will need further evaluation by larger future studies. Meanwhile, we noted that some of our identified features are different from those reported in pre-vious breast MRI-based studies (11). This may be interpreted by the intrinsic difference in the nature of the two-dimensional mammogram images and three-dimensional DCE-MRI. This merits a further analysis and comparison of the two imaging modalities in future work. Because in our study cohort only a very small portion of the patients had breast MRI available, this PS-341 prevented us from performing a comparative study between the two modalities for radiomic-based prediction of breast cancer subtypes.
We compared the effects of using MLO view, CC view, and their combination in this specific subtype classification task. Our finding that the combination of the two views achieved a higher performance may be due to the fact that the two views can provide more information than either of them alone.
Our study has several limitations. First, this was a retrospec-tive analysis of a single-vendor image acquired at a single institution. It will be critical to evaluate whether our findings will generalize on other vendor images and external data. A future multicenter study may help address this question. Sec-ond, we used clinical immunohistochemical surrogate markers to categorize breast cancer subtypes. This is a routine approach in most such studies since in standard clinical 
practice genetic assay is not available to define the subtypes. Third, we have mentioned that we were not able to compare the performance of mammogram images and DCE-MRI in this cohort due to the lack of MRI data. However, we would like to point out that this will be an important study to follow up. In addition, note that we did not separate luminal A and luminal B subtypes in our classification, mainly because they are similar and there is only a small difference between them in clinical management. Finally, we believe catastrophism will be worth-while to test and compare to other machine learning techni-ques for the subtype classification work and we plan to do so in next steps.
Mammogram images are the most commonly available examination for breast cancer screening and diagnosis. If the automated radiomic features like we identified in this study are validated to be predictive of the molecular subtypes, it can provide further information from the images to aid radi-ologists in mammographic reading and to better inform clini-cal diagnosis and decision-making. This would have important additional value too for patients who do not have a breast MRI scan available.
In summary, this pilot radiomics study showed that quanti-tative imaging features extracted from digital mammograms are associated with breast cancer subtypes. Future larger stud-ies are needed to further evaluate the findings and examine the relationship with breast MRI-identified features.
ACKNOWLEDGMENTS
This work was supported by the Key Project of Tianjin Sci-ence and Technology Committee Foundation grant (12ZCDZSY16000) and the Tianjin Municipal Government of China (15JCQNJC14500). This work was also partially supported by a National Institutes of Health (NIH)/National Cancer Institute (NCI) R01 grant (No. 1R01CA193603) and an R01 Supplement grant (No. 3R01CA193603-03S1).