No abstract

Cluster of microcalcifications can be an early sign of breast cancer. In this paper, we propose a novel approach based on convolutional neural networks for the detection and segmentation of microcalcification clusters. In this work, we used 283 mammograms to train and validate our model, obtaining an accuracy of 99.99% on microcalcification detection and a false positive rate of 0.005%. Our results show how deep learning could be an effective tool to effectively support radiologists during mammograms examination.

Coronary artery calcium (CAC) is a significant marker of atherosclerosis and cardiovascular events. In this work we present a system for the automatic quantification of calcium score in ECG-triggered non-contrast enhanced cardiac computed tomography (CT) images. The proposed system uses a supervised deep learning algorithm, i.e. convolutional neural network (CNN) for the segmentation and classification of candidate lesions as coronary or not, previously extracted in the region of the heart using a cardiac atlas. We trained our network with 45 CT volumes; 18 volumes were used to validate the model and 56 to test it. Individual lesions were detected with a sensitivity of 91.24%, a specificity of 95.37% and a positive predicted value (PPV) of 90.5%; comparing calcium score obtained by the system and calcium score manually evaluated by an expert operator, a Pearson coefficient of 0.983 was obtained. A high agreement (Cohenâ\u80\u99s κ = 0.879) between manual and automatic risk prediction was also observed. These results demonstrated that convolutional neural networks can be effectively applied for the automatic segmentation and classification of coronary calcifications

There has been an increasing focus in learning interpretable feature representations, particularly in applications such as medical image analysis that require explainability, whilst relying less on annotated data (since annotations can be tedious and costly). Here we build on recent innovations in style-content representations to learn anatomy, imaging characteristics (appearance) and temporal correlations. By introducing a self-supervised objective of predicting future cardiac phases we improve disentanglement. We propose a temporal transformer architecture that given an image conditioned on phase difference, it predicts a future frame. This forces the anatomical decomposition to be consistent with the temporal cardiac contraction in cine MRI and to have semantic meaning with less need for annotations. We demonstrate that using this regularization, we achieve competitive results and improve semisupervised segmentation, especially when very few labelled data are available. Specifically, we show Dice increase of up to 19% and 7% compared to supervised and semi-supervised approaches respectively on the ACDC dataset. Code is available at: https://github.com/gvalvano/sdtnet.

No abstract

Thanks to their ability to learn flexible data-driven losses, Generative Adversarial Networks (GANs) are an integral part of many semi-and weakly-supervised methods for medical image segmentation. GANs jointly optimise a generator and an adversarial discriminator on a set of training data. After training has completed, the discriminator is usually discarded and only the generator is used for inference. But should we discard discriminators? In this work, we argue that training stable discriminators produces expressive loss functions that we can re-use at inference to detect and correct segmentation mistakes. First, we identify key challenges and suggest possible solutions to make discriminators re-usable at inference. Then, we show that we can combine discriminators with image reconstruction costs (via decoders) to further improve the model. Our method is simple and improves the test-time performance of pre-trained GANs. Moreover, we show that it is compatible with standard post-processing techniques and it has potentials to be used for Online Continual Learning. With our work, we open new research avenues for re-using adversarial discriminators at inference.

Skull-stripping methods aim to remove the non-brain tissue from acquisition of brain scans in magnetic resonance (MR) imaging. Although several methods sharing this common purpose have been presented in literature, they all suffer from the great variability of the MR images. In this work we propose a novel approach based on Convolutional Neural Networks to automatically perform the brain extraction obtaining cutting-edge performance in the NFBS public database. Additionally, we focus on the efficient training of the neural network designing an effective data augmentation pipeline. Obtained results are evaluated through Dice metric, obtaining a value of 96.5%, and processing time, with 4.5s per volume.

No abstract