Christian Federau

Sophie Ostmeier, Brian Axelrod, Benjamin Pulli et al.

Purpose: Multi-expert deep learning training methods to automatically quantify ischemic brain tissue on Non-Contrast CT Materials and Methods: The data set consisted of 260 Non-Contrast CTs from 233 patients of acute ischemic stroke patients recruited in the DEFUSE 3 trial. A benchmark U-Net was trained on the reference annotations of three experienced neuroradiologists to segment ischemic brain tissue using majority vote and random expert sampling training schemes. We used a one-sided Wilcoxon signed-rank test on a set of segmentation metrics to compare bootstrapped point estimates of the training schemes with the inter-expert agreement and ratio of variance for consistency analysis. We further compare volumes with the 24h-follow-up DWI (final infarct core) in the patient subgroup with full reperfusion and we test volumes for correlation to the clinical outcome (mRS after 30 and 90 days) with the Spearman method. Results: Random expert sampling leads to a model that shows better agreement with experts than experts agree among themselves and better agreement than the agreement between experts and a majority-vote model performance (Surface Dice at Tolerance 5mm improvement of 61% to 0.70 +- 0.03 and Dice improvement of 25% to 0.50 +- 0.04). The model-based predicted volume similarly estimated the final infarct volume and correlated better to the clinical outcome than CT perfusion. Conclusion: A model trained on random expert sampling can identify the presence and location of acute ischemic brain tissue on Non-Contrast CT similar to CT perfusion and with better consistency than experts. This may further secure the selection of patients eligible for endovascular treatment in less specialized hospitals.

Christopher Vogelsanger, Christian Federau

Multiple Sclerosis (MS) and microvascular leukoencephalopathy are two distinct neurological conditions, the first caused by focal autoimmune inflammation in the central nervous system, the second caused by chronic white matter damage from atherosclerotic microvascular disease. Both conditions lead to signal anomalies on Fluid Attenuated Inversion Recovery (FLAIR) magnetic resonance (MR) images, which can be distinguished by an expert neuroradiologist, but which can look very similar to the untrained eye as well as in the early stage of both diseases. In this paper, we attempt to train a 3-dimensional deep neural network to learn the specific features of both diseases in an unsupervised manner. For this manner, in a first step we train a generative neural network to create artificial MR images of both conditions with approximate explicit density, using a mixed dataset of multiple sclerosis, leukoencephalopathy and healthy patients containing in total 5404 volumes of 3096 patients. In a second step, we distinguish features between the different diseases in the latent space of this network, and use them to classify new data.

Moritz Platscher, Jonathan Zopes, Christian Federau

Deep learning based disease detection and segmentation algorithms promise to improve many clinical processes. However, such algorithms require vast amounts of annotated training data, which are typically not available in the medical context due to data privacy, legal obstructions, and non-uniform data acquisition protocols. Synthetic databases with annotated pathologies could provide the required amounts of training data. We demonstrate with the example of ischemic stroke that an improvement in lesion segmentation is feasible using deep learning based augmentation. To this end, we train different image-to-image translation models to synthesize magnetic resonance images of brain volumes with and without stroke lesions from semantic segmentation maps. In addition, we train a generative adversarial network to generate synthetic lesion masks. Subsequently, we combine these two components to build a large database of synthetic stroke images. The performance of the various models is evaluated using a U-Net which is trained to segment stroke lesions on a clinical test set. We report a Dice score of $\mathbf{72.8}$% [$\mathbf{70.8\pm1.0}$%] for the model with the best performance, which outperforms the model trained on the clinical images alone $\mathbf{67.3}$% [$\mathbf{63.2\pm1.9}$%], and is close to the human inter-reader Dice score of $\mathbf{76.9}$%. Moreover, we show that for a small database of only 10 or 50 clinical cases, synthetic data augmentation yields significant improvement compared to a setting where no synthetic data is used. To the best of our knowledge, this presents the first comparative analysis of synthetic data augmentation based on image-to-image translation, and first application to ischemic stroke.

Jonathan Zopes, Moritz Platscher, Silvio Paganucci et al.

Purpose: To implement a brain segmentation pipeline based on convolutional neural networks, which rapidly segments 3D volumes into 27 anatomical structures. To provide an extensive, comparative study of segmentation performance on various contrasts of magnetic resonance imaging (MRI) and computed tomography (CT) scans. Methods: Deep convolutional neural networks are trained to segment 3D MRI (MPRAGE, DWI, FLAIR) and CT scans. A large database of in total 851 MRI/CT scans is used for neural network training. Training labels are obtained on the MPRAGE contrast and coregistered to the other imaging modalities. The segmentation quality is quantified using the Dice metric for a total of 27 anatomical structures. Dropout sampling is implemented to identify corrupted input scans or low-quality segmentations. Full segmentation of 3D volumes with more than 2 million voxels is obtained in less than 1s of processing time on a graphical processing unit. Results: The best average Dice score is found on $T_1$-weighted MPRAGE ($85.3\pm4.6\,\%$). However, for FLAIR ($80.0\pm7.1\,\%$), DWI ($78.2\pm7.9\,\%$) and CT ($79.1\pm 7.9\,\%$), good-quality segmentation is feasible for most anatomical structures. Corrupted input volumes or low-quality segmentations can be detected using dropout sampling. Conclusion: The flexibility and performance of deep convolutional neural networks enables the direct, real-time segmentation of FLAIR, DWI and CT scans without requiring $T_1$-weighted scans.

Alejandro Ungría Hirte, Moritz Platscher, Thomas Joyce et al.

We show that high quality, diverse and realistic-looking diffusion-weighted magnetic resonance images can be synthesized using deep generative models. Based on professional neuroradiologists' evaluations and diverse metrics with respect to quality and diversity of the generated synthetic brain images, we present two networks, the Introspective Variational Autoencoder and the Style-Based GAN, that qualify for data augmentation in the medical field, where information is saved in a dispatched and inhomogeneous way and access to it is in many aspects restricted.