We present a novel methodology for the automated detection of breast lesions from dynamic contrast-enhanced magnetic resonance volumes (DCE-MRI). Our method, based on deep reinforcement learning, significantly reduces the inference time for lesion detection compared to an exhaustive search, while retaining state-of-art accuracy.
The video on the left hand side shows a few runs of the RL inference on test MRI volumes.
Complete system for detecting, segmenting and analysing breast masses from mammograms using deep learning models!
Globally optimal inference in a continuous space using a shape prior computed from a semantic segmentation produced by a deep learning model.
Breast screening with Residual Deep Nets.
This work addresses the following question: is it possible to leverage the previously designed hand-crafted features in the modelling of deep convolutional neural networks for the problem of classifying breast masses from mammograms?
This paper explores the use of deep convolution and deep belief networks as potential functions in structured prediction models for the segmentation of breast masses from mammograms
In this work, we propose a precise mass detection approach from mammograms using a cascade of detectors, followed by a random forest classifier.
Use of level set methods to compensate for small training sets to train deep learning models in medical image analysis.
New level set optimisation function that segments jointly multiple overlapping cells.