While the importance of automatic image analysis is increasing at an enormous pace, recent meta-research revealed major flaws with respect to algorithm validation. Specifically, performance metrics are key for objective, transparent and comparative performance assessment, but relatively little attention has been given to the practical pitfalls when using specific metrics for a given image analysis task. A common mission of several international initiatives is therefore to provide researchers with guidelines and tools to choose the performance metrics in a problem-aware manner. This dynamically updated document has the purpose to illustrate important limitations of performance metrics commonly applied in the field of image analysis. The current version is based on a Delphi process on metrics conducted by an international consortium of image analysis experts.
Predicting the final ischaemic stroke lesion provides crucial information regarding the volume of salvageable hypoperfused tissue, which helps physicians in the difficult decision-making process of treatment planning and intervention. Treatment selection is influenced by clinical diagnosis, which requires delineating the stroke lesion, as well as characterising cerebral blood flow dynamics using neuroimaging acquisitions. Nonetheless, predicting the final stroke lesion is an intricate task, due to the variability in lesion size, shape, location and the underlying cerebral haemodynamic processes that occur after the ischaemic stroke takes place. Moreover, since elapsed time between stroke and treatment is related to the loss of brain tissue, assessing and predicting the final stroke lesion needs to be performed in a short period of time, which makes the task even more complex. Therefore, there is a need for automatic methods that predict the final stroke lesion and support physicians in the treatment decision process. We propose a fully automatic deep learning method based on unsupervised and supervised learning to predict the final stroke lesion after 90 days. Our aim is to predict the final stroke lesion location and extent, taking into account the underlying cerebral blood flow dynamics that can influence the prediction. To achieve this, we propose a two-branch Restricted Boltzmann Machine, which provides specialized data-driven features from different sets of standard parametric Magnetic Resonance Imaging maps. These data-driven feature maps are then combined with the parametric Magnetic Resonance Imaging maps, and fed to a Convolutional and Recurrent Neural Network architecture. We evaluated our proposal on the publicly available ISLES 2017 testing dataset, reaching a Dice score of 0.38, Hausdorff Distance of 29.21 mm, and Average Symmetric Surface Distance of 5.52 mm.
Background and Objective: Deep learning enables tremendous progress in medical image analysis. One driving force of this progress are open-source frameworks like TensorFlow and PyTorch. However, these frameworks rarely address issues specific to the domain of medical image analysis, such as 3-D data handling and distance metrics for evaluation. pymia, an open-source Python package, tries to address these issues by providing flexible data handling and evaluation independent of the deep learning framework. Methods: The pymia package provides data handling and evaluation functionalities. The data handling allows flexible medical image handling in every commonly used format (e.g., 2-D, 2.5-D, and 3-D; full- or patch-wise). Even data beyond images like demographics or clinical reports can easily be integrated into deep learning pipelines. The evaluation allows stand-alone result calculation and reporting, as well as performance monitoring during training using a vast amount of domain-specific metrics for segmentation, reconstruction, and regression. Results: The pymia package is highly flexible, allows for fast prototyping, and reduces the burden of implementing data handling routines and evaluation methods. While data handling and evaluation are independent of the deep learning framework used, they can easily be integrated into TensorFlow and PyTorch pipelines. The developed package was successfully used in a variety of research projects for segmentation, reconstruction, and regression. Conclusions: The pymia package fills the gap of current deep learning frameworks regarding data handling and evaluation in medical image analysis. It is available at https://github.com/rundherum/pymia and can directly be installed from the Python Package Index using pip install pymia.
Magnetic resonance fingerprinting (MRF) enables fast and multiparametric MR imaging. Despite fast acquisition, the state-of-the-art reconstruction of MRF based on dictionary matching is slow and lacks scalability. To overcome these limitations, neural network (NN) approaches estimating MR parameters from fingerprints have been proposed recently. Here, we revisit NN-based MRF reconstruction to jointly learn the forward process from MR parameters to fingerprints and the backward process from fingerprints to MR parameters by leveraging invertible neural networks (INNs). As a proof-of-concept, we perform various experiments showing the benefit of learning the forward process, i.e., the Bloch simulations, for improved MR parameter estimation. The benefit especially accentuates when MR parameter estimation is difficult due to MR physical restrictions. Therefore, INNs might be a feasible alternative to the current solely backward-based NNs for MRF reconstruction.
Magnetic resonance fingerprinting (MRF) provides a unique concept for simultaneous and fast acquisition of multiple quantitative MR parameters. Despite acquisition efficiency, adoption of MRF into the clinics is hindered by its dictionary-based reconstruction, which is computationally demanding and lacks scalability. Here, we propose a convolutional neural network-based reconstruction, which enables both accurate and fast reconstruction of parametric maps, and is adaptable based on the needs of spatial regularization and the capacity for the reconstruction. We evaluated the method using MRF T1-FF, an MRF sequence for T1 relaxation time of water and fat fraction mapping. We demonstrate the method's performance on a highly heterogeneous dataset consisting of 164 patients with various neuromuscular diseases imaged at thighs and legs. We empirically show the benefit of incorporating spatial regularization during the reconstruction and demonstrate that the method learns meaningful features from MR physics perspective. Further, we investigate the ability of the method to handle highly heterogeneous morphometric variations and its generalization to anatomical regions unseen during training. The obtained results outperform the state-of-the-art in deep learning-based MRF reconstruction. Coupled with fast MRF sequences, the proposed method has the potential of enabling multiparametric MR imaging in clinically feasible time.
We propose a Dual-Stream Pyramid Registration Network (referred as Dual-PRNet) for unsupervised 3D medical image registration. Unlike recent CNN-based registration approaches, such as VoxelMorph, which explores a single-stream encoder-decoder network to compute a registration fields from a pair of 3D volumes, we design a two-stream architecture able to compute multi-scale registration fields from convolutional feature pyramids. Our contributions are two-fold: (i) we design a two-stream 3D encoder-decoder network which computes two convolutional feature pyramids separately for a pair of input volumes, resulting in strong deep representations that are meaningful for deformation estimation; (ii) we propose a pyramid registration module able to predict multi-scale registration fields directly from the decoding feature pyramids. This allows it to refine the registration fields gradually in a coarse-to-fine manner via sequential warping, and enable the model with the capability for handling significant deformations between two volumes, such as large displacements in spatial domain or slice space. The proposed Dual-PRNet is evaluated on two standard benchmarks for brain MRI registration, where it outperforms the state-of-the-art approaches by a large margin, e.g., having improvements over recent VoxelMorph [2] with 0.683->0.778 on the LPBA40, and 0.511->0.631 on the Mindboggle101, in term of average Dice score.
Despite the recent improvements in overall accuracy, deep learning systems still exhibit low levels of robustness. Detecting possible failures is critical for a successful clinical integration of these systems, where each data point corresponds to an individual patient. Uncertainty measures are a promising direction to improve failure detection since they provide a measure of a system's confidence. Although many uncertainty estimation methods have been proposed for deep learning, little is known on their benefits and current challenges for medical image segmentation. Therefore, we report results of evaluating common voxel-wise uncertainty measures with respect to their reliability, and limitations on two medical image segmentation datasets. Results show that current uncertainty methods perform similarly and although they are well-calibrated at the dataset level, they tend to be miscalibrated at subject-level. Therefore, the reliability of uncertainty estimates is compromised, highlighting the importance of developing subject-wise uncertainty estimations. Additionally, among the benchmarked methods, we found auxiliary networks to be a valid alternative to common uncertainty methods since they can be applied to any previously trained segmentation model.
Volumetric image segmentation with convolutional neural networks (CNNs) encounters several challenges, which are specific to medical images. Among these challenges are large volumes of interest, high class imbalances, and difficulties in learning shape representations. To tackle these challenges, we propose to improve over traditional CNN-based volumetric image segmentation through point-wise classification of point clouds. The sparsity of point clouds allows processing of entire image volumes, balancing highly imbalanced segmentation problems, and explicitly learning an anatomical shape. We build upon PointCNN, a neural network proposed to process point clouds, and propose here to jointly encode shape and volumetric information within the point cloud in a compact and computationally effective manner. We demonstrate how this approach can then be used to refine CNN-based segmentation, which yields significantly improved results in our experiments on the difficult task of peripheral nerve segmentation from magnetic resonance neurography images. By synthetic experiments, we further show the capability of our approach in learning an explicit anatomical shape representation.
Training robust deep learning (DL) systems for disease detection from medical images is challenging due to limited images covering different disease types and severity. The problem is especially acute, where there is a severe class imbalance. We propose an active learning (AL) framework to select most informative samples for training our model using a Bayesian neural network. Informative samples are then used within a novel class aware generative adversarial network (CAGAN) to generate realistic chest xray images for data augmentation by transferring characteristics from one class label to another. Experiments show our proposed AL framework is able to achieve state-of-the-art performance by using about $35\%$ of the full dataset, thus saving significant time and effort over conventional methods.