Machine learning in medical imaging. MLMI (Workshop)最新文献

Current spherical surface registration methods achieve good performance on alignment and spatial normalization of cortical surfaces across individuals in neuroimaging analysis. However, they are computationally intensive, since they have to optimize an objective function independently for each pair of surfaces. In this paper, we present a fast learning-based algorithm that makes use of the recent development in spherical Convolutional Neural Networks (CNNs) for spherical cortical surface registration. Given a set of surface pairs without supervised information such as ground truth deformation fields or anatomical landmarks, we formulate the registration as a parametric function and learn its parameters by enforcing the feature similarity between one surface and the other one warped by the estimated deformation field using the function. Then, given a new pair of surfaces, we can quickly infer the spherical deformation field registering one surface to the other one. We model this parametric function using three orthogonal Spherical U-Nets and use spherical transform layers to warp the spherical surfaces, while imposing smoothness constraints on the deformation field. All the layers in the network are well-defined and differentiable, thus the parameters can be effectively learned. We show that our method achieves accurate cortical alignment results on 102 subjects, comparable to two state-of-the-art methods: Spherical Demons and MSM, while runs much faster.

Registration of images with pathologies is challenging due to tissue appearance changes and missing correspondences caused by the pathologies. Moreover, mass effects as observed for brain tumors may displace tissue, creating larger deformations over time than what is observed in a healthy brain. Deep learning models have successfully been applied to image registration to offer dramatic speed up and to use surrogate information (e.g., segmentations) during training. However, existing approaches focus on learning registration models using images from healthy patients. They are therefore not designed for the registration of images with strong pathologies for example in the context of brain tumors, and traumatic brain injuries. In this work, we explore a deep learning approach to register images with brain tumors to an atlas. Our model learns an appearance mapping from images with tumors to the atlas, while simultaneously predicting the transformation to atlas space. Using separate decoders, the network disentangles the tumor mass effect from the reconstruction of quasi-normal images. Results on both synthetic and real brain tumor scans show that our approach outperforms cost function masking for registration to the atlas and that reconstructed quasi-normal images can be used for better longitudinal registrations.

Deep learning strategies have become ubiquitous optimization tools for medical image analysis. With the appropriate amount of data, these approaches outperform classic methodologies in a variety of image processing tasks. However, rare diseases and pediatric imaging often lack extensive data. Specially, MRI are uncommon because they require sedation in young children. Moreover, the lack of standardization in MRI protocols introduces a strong variability between different datasets. In this paper, we present a general deep learning architecture for MRI homogenization that also provides the segmentation map of an anatomical region of interest. Homogenization is achieved using an unsupervised architecture based on variational autoencoder with cycle generative adversarial networks, which learns a common space (i.e. a representation of the optimal imaging protocol) using an unpaired image-to-image translation network. The segmentation is simultaneously generated by a supervised learning strategy. We evaluated our method segmenting the challenging anterior visual pathway using three brain T1-weighted MRI datasets (variable protocols and vendors). Our method significantly outperformed a non-homogenized multi-protocol U-Net.

Traditional intensity-based 2D/3D registration requires near-perfect initialization in order for image similarity metrics to yield meaningful updates of X-ray pose and reduce the likelihood of getting trapped in a local minimum. The conventional approaches strongly depend on image appearance rather than content, and therefore, fail in revealing large pose offsets that substantially alter the appearance of the same structure. We complement traditional similarity metrics with a convolutional neural network-based (CNN-based) registration solution that captures large-range pose relations by extracting both local and contextual information, yielding meaningful X-ray pose updates without the need for accurate initialization. To register a 2D X-ray image and a 3D CT scan, our CNN accepts a target X-ray image and a digitally reconstructed radiograph at the current pose estimate as input and iteratively outputs pose updates in the direction of the pose gradient on the Riemannian Manifold. Our approach integrates seamlessly with conventional image-based registration frameworks, where long-range relations are captured primarily by our CNN-based method while short-range offsets are recovered accurately with an image similarity-based method. On both synthetic and real X-ray images of the human pelvis, we demonstrate that the proposed method can successfully recover large rotational and translational offsets, irrespective of initialization.

Convolutional neural networks (CNNs) have recently been popular for classification and segmentation through numerous network architectures offering a substantial performance improvement. Their value has been particularly appreciated in the domain of biomedical applications, where even a small improvement in the predicted segmented region (e.g., a malignancy) compared to the ground truth can potentially lead to better diagnosis or treatment planning. Here, we introduce a novel architecture, namely the Overall Convolutional Network (O-Net), which takes advantage of different pooling levels and convolutional layers to extract more deeper local and containing global context. Our quantitative results on 2D images from two distinct datasets show that O-Net can achieve a higher dice coefficient when compared to either a U-Net or a Pyramid Scene Parsing Net. We also look into the stability of results for training and validation sets which can show the robustness of model compared with new datasets. In addition to comparison to the decoder, we use different encoders including simple, VGG Net, and ResNet. The ResNet encoder could help to improve the results in most of the cases.

Heterogeneous presentation of a neurological disorder suggests potential differences in the underlying pathophysiological changes that occur in the brain. We propose to model heterogeneous patterns of functional network differences using a demographic-guided attention (DGA) mechanism for recurrent neural network models for prediction from functional magnetic resonance imaging (fMRI) time-series data. The context computed from the DGA head is used to help focus on the appropriate functional networks based on individual demographic information. We demonstrate improved classification on 3 subsets of the ABIDE I dataset used in published studies that have previously produced state-of-the-art results, evaluating performance under a leave-one-site-out cross-validation framework for better generalizeability to new data. Finally, we provide examples of interpreting functional network differences based on individual demographic variables.

The connectivity analysis is a powerful technique for investigating a hard-wired brain architecture as well as flexible, functional dynamics tied to human cognition. Recent multi-modal connectivity studies had the challenge of combining functional and structural connectivity information into one integrated network. In this paper, we proposed a simplex regression model with graph-constrained Elastic Net (GraphNet) to estimate functional networks enriched by structural connectivity in a biologically meaningful way with a low model complexity. Our model constructed the functional networks using sparse simplex regression framework and enriched structural connectivity information based on GraphNet constraint. We applied our model on the real neuroimaging datasets to show its ability for predicting a clinical score. Our results demonstrated that integrating multi-modal features could detect more sensitive and subtle brain biomarkers than using a single modality.

With recent advances in deep learning, neuroimaging studies increasingly rely on convolutional networks (ConvNets) to predict diagnosis based on MR images. To gain a better understanding of how a disease impacts the brain, the studies visualize the salience maps of the ConvNet highlighting voxels within the brain majorly contributing to the prediction. However, these salience maps are generally confounded, i.e., some salient regions are more predictive of confounding variables (such as age) than the diagnosis. To avoid such misinterpretation, we propose in this paper an approach that aims to visualize confounder-free saliency maps that only highlight voxels predictive of the diagnosis. The approach incorporates univariate statistical tests to identify confounding effects within the intermediate features learned by ConvNet. The influence from the subset of confounded features is then removed by a novel partial back-propagation procedure. We use this two-step approach to visualize confounder-free saliency maps extracted from synthetic and two real datasets. These experiments reveal the potential of our visualization in producing unbiased model-interpretation.