SUMMARY: Python is the most commonly used language for deep learning (DL). Existing Python packages for mass spectrometry imaging (MSI) data are not optimized for DL tasks. We, therefore, introduce pyM2aia, a Python package for MSI data analysis with a focus on memory-efficient handling, processing and convenient data-access for DL applications. pyM2aia provides interfaces to its parent application M2aia, which offers interactive capabilities for exploring and annotating MSI data in imzML format. pyM2aia utilizes the image input and output routines, data formats, and processing functions of M2aia, ensures data interchangeability, and enables the writing of readable and easy-to-maintain DL pipelines by providing batch generators for typical MSI data access strategies. We showcase the package in several examples, including imzML metadata parsing, signal processing, ion-image generation, and, in particular, DL model training and inference for spectrum-wise approaches, ion-image-based approaches, and approaches that use spectral and spatial information simultaneously. AVAILABILITY AND IMPLEMENTATION: Python package, code and examples are available at (https://m2aia.github.io/m2aia).

Anisotropic multi-slice Cardiac Magnetic Resonance (CMR) Images are conventionally acquired in patient-specific short-axis (SAX) orientation. In specific cardiovascular diseases that affect right ventricular (RV) morphology, acquisitions in standard axial (AX) orientation are preferred by some investigators, due to potential superiority in RV volume measurement for treatment planning. Unfortunately, due to the rare occurrence of these diseases, data in this domain is scarce. Recent research in deep learning-based methods mainly focused on SAX CMR images and they had proven to be very successful. In this work, we show that there is a considerable domain shift between AX and SAX images, and therefore, direct application of existing models yield sub-optimal results on AX samples. We propose a novel unsupervised domain adaptation approach, which uses task-related probabilities in an attention mechanism. Beyond that, cycle consistency is imposed on the learned patient-individual 3D rigid transformation to improve stability when automatically re-sampling the AX images to SAX orientations. The network was trained on 122 registered 3D AX-SAX CMR volume pairs from a multi-centric patient cohort. A mean 3D Dice of 0.86 ± 0.06 for the left ventricle, 0.65 ± 0.08 for the myocardium, and 0.77 ± 0.10 for the right ventricle could be achieved. This is an improvement of 25% in Dice for RV in comparison to direct application on axial slices. To conclude, our pre-trained task module has neither seen CMR images nor labels from the target domain, but is able to segment them after the domain gap is reduced. Code: https://github.com/Cardio-AI/3d-mri-domain-adaptation.

Delineation of the left ventricular cavity, myocardium, and right ventricle from cardiac magnetic resonance images (multi-slice 2-D cine MRI) is a common clinical task to establish diagnosis. The automation of the corresponding tasks has thus been the subject of intense research over the past decades. In this paper, we introduce the "Automatic Cardiac Diagnosis Challenge" dataset (ACDC), the largest publicly available and fully annotated dataset for the purpose of cardiac MRI (CMR) assessment. The dataset contains data from 150 multi-equipments CMRI recordings with reference measurements and classification from two medical experts. The overarching objective of this paper is to measure how far state-of-the-art deep learning methods can go at assessing CMRI, i.e., segmenting the myocardium and the two ventricles as well as classifying pathologies. In the wake of the 2017 MICCAI-ACDC challenge, we report results from deep learning methods provided by nine research groups for the segmentation task and four groups for the classification task. Results show that the best methods faithfully reproduce the expert analysis, leading to a mean value of 0.97 correlation score for the automatic extraction of clinical indices and an accuracy of 0.96 for automatic diagnosis. These results clearly open the door to highly accurate and fully automatic analysis of cardiac CMRI. We also identify scenarios for which deep learning methods are still failing. Both the dataset and detailed results are publicly available online, while the platform will remain open for new submissions.

This paper presents a novel end-to-end system for table understanding in document images called DeepDeSRT. In particular, the contribution of DeepDeSRT is two-fold. First, it presents a deep learning-based solution for table detection in document images. Secondly, it proposes a novel deep learning-based approach for table structure recognition, i.e. identifying rows, columns, and cell positions in the detected tables. In contrast to existing rule-based methods, which rely on heuristics or additional PDF metadata (like, for example, print instructions, character bounding boxes, or line segments), the presented system is data-driven and does not need any heuristics or metadata to detect as well as to recognize tabular structures in document images. Furthermore, in contrast to most existing table detection and structure recognition methods, which are applicable only to PDFs, DeepDeSRT processes document images, which makes it equally suitable for born-digital PDFs (as they can automatically be converted into images) as well as even harder problems, e.g. scanned documents. To gauge the performance of DeepDeSRT, the system is evaluated on the publicly available ICDAR 2013 table competition dataset containing 67 documents with 238 pages overall. Evaluation results reveal that DeepDeSRT outperforms state-of-the-art methods for table detection and structure recognition and achieves F1-measures of 96.77% and 91.44% for table detection and structure recognition, respectively. Additionally, DeepDeSRT is evaluated on a closed dataset from a real use case of a major European aviation company comprising documents which are highly unlike those in ICDAR 2013. Tested on a randomly selected sample from this dataset, DeepDeSRT achieves high detection accuracy for tables which demonstrates the sound generalization capabilities of our system.

Effective image-based artifact correction is an essential step in the analysis of diffusion MR images. Many current approaches are based on retrospective registration, which becomes challenging in the realm of high b-values and low signal-to-noise ratio, rendering the corresponding correction schemes more and more ineffective. We propose a novel registration scheme based on memetic search optimization that allows for simultaneous exploitation of different signal intensity relationships between the images, leading to more robust registration results. We demonstrate the increased robustness and efficacy of our method on simulated as well as in vivo datasets. In contrast to the state-of-art methods, the median target registration error (TRE) stayed below the voxel size even for high b-values (3000 s · mm-2and higher) and low SNR conditions. We also demonstrate the increased precision in diffusion-derived quantities by evaluating Neurite Orientation Dispersion and Density Imaging (NODDI) derived measures on a in vivo dataset with severe motion artifacts. These promising results will potentially inspire further studies on metaheuristic optimization in diffusion MRI artifact correction and image registration in general.

This paper presents a comparison study between 10 automatic and six interactive methods for liver segmentation from contrast-enhanced CT images. It is based on results from the "MICCAI 2007 Grand Challenge" workshop, where 16 teams evaluated their algorithms on a common database. A collection of 20 clinical images with reference segmentations was provided to train and tune algorithms in advance. Participants were also allowed to use additional proprietary training data for that purpose. All teams then had to apply their methods to 10 test datasets and submit the obtained results. Employed algorithms include statistical shape models, atlas registration, level-sets, graph-cuts and rule-based systems. All results were compared to reference segmentations five error measures that highlight different aspects of segmentation accuracy. All measures were combined according to a specific scoring system relating the obtained values to human expert variability. In general, interactive methods reached higher average scores than automatic approaches and featured a better consistency of segmentation quality. However, the best automatic methods (mainly based on statistical shape models with some additional free deformation) could compete well on the majority of test images. The study provides an insight in performance of different segmentation approaches under real-world conditions and highlights achievements and limitations of current image analysis techniques.

Most image processing and visualization applications allow users to configure computation parameters and manipulate the resulting visualizations. SCIRun, VolView, MeVisLab, and the Medical Interaction Toolkit (MITK) are four image processing and visualization frameworks that were built for these purposes. All frameworks are freely available and all allow the use of the ITK C++ library. In this paper, the benefits and limitations of each visualization framework are presented to aid both application developers and users in the decision of which framework may be best to use for their application. The analysis is based on more than 50 evaluation criteria, functionalities, and example applications. We report implementation times for various steps in the creation of a reference application in each of the compared frameworks. The data-flow programming frameworks, SCIRun and MeVisLab, were determined to be best for developing application prototypes, while VolView was advantageous for nonautomatic end-user applications based on existing ITK functionalities, and MITK was preferable for automated end-user applications that might include new ITK classes specifically designed for the application.

Echocardiography (cardiac ultrasound) is today the predominant technique for quantitative assessment of cardiac function and valvular heart lesions. Segmentation of cardiac structures is required to determine many important diagnostic parameters. As the heart is a moving organ, reliable information can be obtained only from three-dimensional (3-D) data over time (3-D + time = 4-D). Due to their size, the resulting four-dimensional (4-D) data sets are not reasonably accessible to simple manual segmentation methods. Automatic segmentation often yields unsatisfactory results in a clinical environment, especially for ultrasonic images. We describe a semiautomated segmentation algorithm (ROPES) that is able to greatly reduce the time necessary for user interaction and its application to extract various parameters from 4-D echocardiographic data. After searching for candidate contour points, which have to fulfill a multiscale edge criterion, the candidates are connected by minimizing a cost function to line segments that then are connected to form a closed contour. The contour is automatically checked for plausibility. If necessary, two correction methods that can also be used interactively are applied (fitting of other line segments into the contour and searching for additional candidates with a relaxed criterion). The method is validated using in vivo transesophageal echocardiographic data sets.