PHiSeg: Capturing Uncertainty in Medical Image Segmentation

Segmentation of anatomical structures and pathologies is inherently ambiguous. For instance, structure borders may not be clearly visible or different experts may have different styles of annotating. The majority of current state-of-the-art methods do not account for such ambiguities but rather learn a single mapping from image to segmentation. In this work, we propose a novel method to model the conditional probability distribution of the segmentations given an input image. We derive a hierarchical probabilistic model, in which separate latent spaces are responsible for modelling the segmentation at different resolutions. Inference in this model can be efficiently performed using the variational autoencoder framework. We show that our proposed method can be used to generate significantly more realistic and diverse segmentation samples compared to recent related work, both, when trained with annotations from a single or multiple annotators.

Semi-Supervised and Task-Driven Data Augmentation

Supervised deep learning methods for segmentation require large amounts of labelled training data, without which they are prone to overfitting, not generalizing well to unseen images. In practice, obtaining a large number of annotations from clinical experts is expensive and time-consuming. One way to address scarcity of annotated examples is data augmentation using random spatial and intensity transformations. Recently, it has been proposed to use generative models to synthesize realistic training examples, complementing the random augmentation. So far, these methods have yielded limited gains over the random augmentation. However, there is potential to improve the approach by (i) explicitly modeling deformation fields (non-affine spatial transformation) and intensity transformations and (ii) leveraging unlabelled data during the generative process. With this motivation, we propose a novel task-driven data augmentation method where to synthesize new training examples, a generative network explicitly models and applies deformation fields and additive intensity masks on existing labelled data, modeling shape and intensity variations, respectively. Crucially, the generative model is optimized to be conducive to the task, in this case segmentation, and constrained to match the distribution of images observed from labelled and unlabelled samples. Furthermore, explicit modeling of deformation fields allow synthesizing segmentation masks and images in exact correspondence by simply applying the generated transformation to an input image and the corresponding annotation. Our experiments on cardiac magnetic resonance images (MRI) showed that, for the task of segmentation in small training data scenarios, the proposed method substantially outperforms conventional augmentation techniques.

Adversarial Augmentation for Enhancing Classification of Mammography Images

Supervised deep learning relies on the assumption that enough training data is available, which presents a problem for its application to several fields, like medical imaging. On the example of a binary image classification task (breast cancer recognition), we show that pretraining a generative model for meaningful image augmentation helps enhance the performance of the resulting classifier. By augmenting the data, performance on downstream classification tasks could be improved even with a relatively small training set. We show that this "adversarial augmentation" yields promising results compared to classical image augmentation on the example of breast cancer classification.

Injecting and removing malignant features in mammography with CycleGAN: Investigation of an automated adversarial attack using neural networks

$\textbf{Purpose}$ To train a cycle-consistent generative adversarial network (CycleGAN) on mammographic data to inject or remove features of malignancy, and to determine whether these AI-mediated attacks can be detected by radiologists. $\textbf{Material and Methods}$ From the two publicly available datasets, BCDR and INbreast, we selected images from cancer patients and healthy controls. An internal dataset served as test data, withheld during training. We ran two experiments training CycleGAN on low and higher resolution images ($256 \times 256$ px and $512 \times 408$ px). Three radiologists read the images and rated the likelihood of malignancy on a scale from 1-5 and the likelihood of the image being manipulated. The readout was evaluated by ROC analysis (Area under the ROC curve = AUC). $\textbf{Results}$ At the lower resolution, only one radiologist exhibited markedly lower detection of cancer (AUC=0.85 vs 0.63, p=0.06), while the other two were unaffected (0.67 vs. 0.69 and 0.75 vs. 0.77, p=0.55). Only one radiologist could discriminate between original and modified images slightly better than guessing/chance (0.66, p=0.008). At the higher resolution, all radiologists showed significantly lower detection rate of cancer in the modified images (0.77-0.84 vs. 0.59-0.69, p=0.008), however, they were now able to reliably detect modified images due to better visibility of artifacts (0.92, 0.92 and 0.97). $\textbf{Conclusion}$ A CycleGAN can implicitly learn malignant features and inject or remove them so that a substantial proportion of small mammographic images would consequently be misdiagnosed. At higher resolutions, however, the method is currently limited and has a clear trade-off between manipulation of images and introduction of artifacts.

Morpho-MNIST: Quantitative Assessment and Diagnostics for Representation Learning

Revealing latent structure in data is an active field of research, having brought exciting new models such as variational autoencoders and generative adversarial networks, and is essential to push machine learning towards unsupervised knowledge discovery. However, a major challenge is the lack of suitable benchmarks for an objective and quantitative evaluation of learned representations. To address this issue we introduce Morpho-MNIST. We extend the popular MNIST dataset by adding a morphometric analysis enabling quantitative comparison of different models, identification of the roles of latent variables, and characterisation of sample diversity. We further propose a set of quantifiable perturbations to assess the performance of unsupervised and supervised methods on challenging tasks such as outlier detection and domain adaptation.

Uncertainty Quantification in CNN-Based Surface Prediction Using Shape Priors

Surface reconstruction is a vital tool in a wide range of areas of medical image analysis and clinical research. Despite the fact that many methods have proposed solutions to the reconstruction problem, most, due to their deterministic nature, do not directly address the issue of quantifying uncertainty associated with their predictions. We remedy this by proposing a novel probabilistic deep learning approach capable of simultaneous surface reconstruction and associated uncertainty prediction. The method incorporates prior shape information in the form of a principal component analysis (PCA) model. Experiments using the UK Biobank data show that our probabilistic approach outperforms an analogous deterministic PCA-based method in the task of 2D organ delineation and quantifies uncertainty by formulating distributions over predicted surface vertex positions.

Combining Heterogeneously Labeled Datasets For Training Segmentation Networks

Accurate segmentation of medical images is an important step towards analyzing and tracking disease related morphological alterations in the anatomy. Convolutional neural networks (CNNs) have recently emerged as a powerful tool for many segmentation tasks in medical imaging. The performance of CNNs strongly depends on the size of the training data and combining data from different sources is an effective strategy for obtaining larger training datasets. However, this is often challenged by heterogeneous labeling of the datasets. For instance, one of the dataset may be missing labels or a number of labels may have been combined into a super label. In this work we propose a cost function which allows integration of multiple datasets with heterogeneous label subsets into a joint training. We evaluated the performance of this strategy on thigh MR and a cardiac MR datasets in which we artificially merged labels for half of the data. We found the proposed cost function substantially outperforms a naive masking approach, obtaining results very close to using the full annotations.

Iterative Interaction Training for Segmentation Editing Networks

Automatic segmentation has great potential to facilitate morphological measurements while simultaneously increasing efficiency. Nevertheless often users want to edit the segmentation to their own needs and will need different tools for this. There has been methods developed to edit segmentations of automatic methods based on the user input, primarily for binary segmentations. Here however, we present an unique training strategy for convolutional neural networks (CNNs) trained on top of an automatic method to enable interactive segmentation editing that is not limited to binary segmentation. By utilizing a robot-user during training, we closely mimic realistic use cases to achieve optimal editing performance. In addition, we show that an increase of the iterative interactions during the training process up to ten improves the segmentation editing performance substantially. Furthermore, we compare our segmentation editing CNN (interCNN) to state-of-the-art interactive segmentation algorithms and show a superior or on par performance.

Generative Adversarial Networks for MR-CT Deformable Image Registration

Deformable Image Registration (DIR) of MR and CT images is one of the most challenging registration task, due to the inherent structural differences of the modalities and the missing dense ground truth. Recently cycle Generative Adversarial Networks (cycle-GANs) have been used to learn the intensity relationship between these 2 modalities for unpaired brain data. Yet its usefulness for DIR was not assessed. In this study we evaluate the DIR performance for thoracic and abdominal organs after synthesis by cycle-GAN. We show that geometric changes, which differentiate the two populations (e.g. inhale vs. exhale), are readily synthesized as well. This causes substantial problems for any application which relies on spatial correspondences being preserved between the real and the synthesized image (e.g. plan, segmentation, landmark propagation). To alleviate this problem, we investigated reducing the spatial information provided to the discriminator by decreasing the size of its receptive fields. Image synthesis was learned from 17 unpaired subjects per modality. Registration performance was evaluated with respect to manual segmentations of 11 structures for 3 subjects from the VISERAL challenge. State-of-the-art DIR methods based on Normalized Mutual Information (NMI), Modality Independent Neighborhood Descriptor (MIND) and their novel combination achieved a mean segmentation overlap ratio of 76.7, 67.7, 76.9%, respectively. This dropped to 69.1% or less when registering images synthesized by cycle-GAN based on local correlation, due to the poor performance on the thoracic region, where large lung volume changes were synthesized. Performance for the abdominal region was similar to that of CT-MRI NMI registration (77.4 vs. 78.8%) when using 3D synthesizing MRIs (12 slices) and medium sized receptive fields for the discriminator.