For the image-only DL model, we implemented a deep convolutional neural network (ResNet18 [13]) with PyTorch (version 0.31; pytorch.org). Given a 1664 × 2048 pixel view of a breast, the DL model was trained to predict whether or not that breast would develop breast cancer within 5 years.
The only “innovation” here is feeding full view mammograms to a ResNet18(2016 model). The traditional risk factors regression is nothing special (barely machine learning). They don’t go in depth about how they combine the two for the hybrid model, so it’s probably safe to assume it is something simple (merely combining the results, so nothing special in the training step). edit: I stand corrected, commenter below pointed out the appendix, and the regression does in fact come into play in the training step
As a different commenter mentioned, the data collection is largely the interesting part here.
I’ll admit I was wrong about my first guess as to the network topology used though, I was thinking they used something like auto encoders (but that is mostly used in cases where examples of bad samples are rare)
They don’t go in depth about how they combine the two for the hybrid model
Actually they did, it’s in Appendix E (PDF warning)
. A GitHub repo would have been nice, but I think there would be enough info to replicate this if we had the data.
Yeah it’s not the most interesting paper in the world. But it’s still a cool use IMO even if it might not be novel enough to deserve a news article.
The only “innovation” here is feeding full view mammograms to a ResNet18(2016 model). The traditional risk factors regression is nothing special (barely machine learning). They don’t go in depth about how they combine the two for the hybrid model,
so it’s probably safe to assume it is something simple (merely combining the results, so nothing special in the training step).edit: I stand corrected, commenter below pointed out the appendix, and the regression does in fact come into play in the training stepAs a different commenter mentioned, the data collection is largely the interesting part here.
I’ll admit I was wrong about my first guess as to the network topology used though, I was thinking they used something like auto encoders (but that is mostly used in cases where examples of bad samples are rare)
Actually they did, it’s in Appendix E (PDF warning) . A GitHub repo would have been nice, but I think there would be enough info to replicate this if we had the data.
Yeah it’s not the most interesting paper in the world. But it’s still a cool use IMO even if it might not be novel enough to deserve a news article.
ResNet18 is ancient and tiny…I don’t understand why they didn’t go with a deeper network. ResNet50 is usually the smallest I’ll use.