Jan S. Kirschke

Julian McGinnis, Suprosanna Shit, Hongwei Bran Li et al.

Clinical routine and retrospective cohorts commonly include multi-parametric Magnetic Resonance Imaging; however, they are mostly acquired in different anisotropic 2D views due to signal-to-noise-ratio and scan-time constraints. Thus acquired views suffer from poor out-of-plane resolution and affect downstream volumetric image analysis that typically requires isotropic 3D scans. Combining different views of multi-contrast scans into high-resolution isotropic 3D scans is challenging due to the lack of a large training cohort, which calls for a subject-specific framework. This work proposes a novel solution to this problem leveraging Implicit Neural Representations (INR). Our proposed INR jointly learns two different contrasts of complementary views in a continuous spatial function and benefits from exchanging anatomical information between them. Trained within minutes on a single commodity GPU, our model provides realistic super-resolution across different pairs of contrasts in our experiments with three datasets. Using Mutual Information (MI) as a metric, we find that our model converges to an optimum MI amongst sequences, achieving anatomically faithful reconstruction. Code is available at: https://github.com/jqmcginnis/multi_contrast_inr/

Qingqiao Hu, Hao Wang, Jing Luo et al.

Automated medical image segmentation inherently involves a certain degree of uncertainty. One key factor contributing to this uncertainty is the ambiguity that can arise in determining the boundaries of a target region of interest, primarily due to variations in image appearance. On top of this, even among experts in the field, different opinions can emerge regarding the precise definition of specific anatomical structures. This work specifically addresses the modeling of segmentation uncertainty, known as inter-rater uncertainty. Its primary objective is to explore and analyze the variability in segmentation outcomes that can occur when multiple experts in medical imaging interpret and annotate the same images. We introduce a novel Bayesian neural network-based architecture to estimate inter-rater uncertainty in medical image segmentation. Our approach has three key advancements. Firstly, we introduce a one-encoder-multi-decoder architecture specifically tailored for uncertainty estimation, enabling us to capture the rater-specific representation of each expert involved. Secondly, we propose Bayesian modeling for the new architecture, allowing efficient capture of the inter-rater distribution, particularly in scenarios with limited annotations. Lastly, we enhance the rater-specific representation by integrating an attention module into each decoder. This module facilitates focused and refined segmentation results for each rater. We conduct extensive evaluations using synthetic and real-world datasets to validate our technical innovations rigorously. Our method surpasses existing baseline methods in five out of seven diverse tasks on the publicly available \emph{QUBIQ} dataset, considering two evaluation metrics encompassing different uncertainty aspects. Our codes, models, and the new dataset are available through our GitHub repository: https://github.com/HaoWang420/bOEMD-net .

Matan Atad, David Schinz, Hendrik Moeller et al.

Counterfactual explanations (CEs) aim to enhance the interpretability of machine learning models by illustrating how alterations in input features would affect the resulting predictions. Common CE approaches require an additional model and are typically constrained to binary counterfactuals. In contrast, we propose a novel method that operates directly on the latent space of a generative model, specifically a Diffusion Autoencoder (DAE). This approach offers inherent interpretability by enabling the generation of CEs and the continuous visualization of the model's internal representation across decision boundaries. Our method leverages the DAE's ability to encode images into a semantically rich latent space in an unsupervised manner, eliminating the need for labeled data or separate feature extraction models. We show that these latent representations are helpful for medical condition classification and the ordinal regression of severity pathologies, such as vertebral compression fractures (VCF) and diabetic retinopathy (DR). Beyond binary CEs, our method supports the visualization of ordinal CEs using a linear model, providing deeper insights into the model's decision-making process and enhancing interpretability. Experiments across various medical imaging datasets demonstrate the method's advantages in interpretability and versatility. The linear manifold of the DAE's latent space allows for meaningful interpolation and manipulation, making it a powerful tool for exploring medical image properties. Our code is available at https://doi.org/10.5281/zenodo.13859266.

Moritz Roman Hernandez Petzsche, Ezequiel de la Rosa, Uta Hanning et al.

Magnetic resonance imaging (MRI) is a central modality for stroke imaging. It is used upon patient admission to make treatment decisions such as selecting patients for intravenous thrombolysis or endovascular therapy. MRI is later used in the duration of hospital stay to predict outcome by visualizing infarct core size and location. Furthermore, it may be used to characterize stroke etiology, e.g. differentiation between (cardio)-embolic and non-embolic stroke. Computer based automated medical image processing is increasingly finding its way into clinical routine. Previous iterations of the Ischemic Stroke Lesion Segmentation (ISLES) challenge have aided in the generation of identifying benchmark methods for acute and sub-acute ischemic stroke lesion segmentation. Here we introduce an expert-annotated, multicenter MRI dataset for segmentation of acute to subacute stroke lesions. This dataset comprises 400 multi-vendor MRI cases with high variability in stroke lesion size, quantity and location. It is split into a training dataset of n=250 and a test dataset of n=150. All training data will be made publicly available. The test dataset will be used for model validation only and will not be released to the public. This dataset serves as the foundation of the ISLES 2022 challenge with the goal of finding algorithmic methods to enable the development and benchmarking of robust and accurate segmentation algorithms for ischemic stroke.

Hendrik Möller, Hanna Schoen, Robert Graf et al.

The human spine commonly consists of seven cervical, twelve thoracic, and five lumbar vertebrae. However, enumeration anomalies may result in individuals having eleven or thirteen thoracic vertebrae and four or six lumbar vertebrae. Although the identification of enumeration anomalies has potential clinical implications for chronic back pain and operation planning, the thoracolumbar junction is often poorly assessed and rarely described in clinical reports. Additionally, even though multiple deep-learning-based vertebra labeling algorithms exist, there is a lack of methods to automatically label enumeration anomalies. Our work closes that gap by introducing "Vertebra Identification with Anomaly Handling" (VERIDAH), a novel vertebra labeling algorithm based on multiple classification heads combined with a weighted vertebra sequence prediction algorithm. We show that our approach surpasses existing models on T2w TSE sagittal (98.30% vs. 94.24% of subjects with all vertebrae correctly labeled, p < 0.001) and CT imaging (99.18% vs. 77.26% of subjects with all vertebrae correctly labeled, p < 0.001) and works in arbitrary field-of-view images. VERIDAH correctly labeled the presence 2 Möller et al. of thoracic enumeration anomalies in 87.80% and 96.30% of T2w and CT images, respectively, and lumbar enumeration anomalies in 94.48% and 97.22% for T2w and CT, respectively. Our code and models are available at: https://github.com/Hendrik-code/spineps.

Alina F. Dima, Veronika A. Zimmer, Martin J. Menten et al.

Automated segmentation of the blood vessels in 3D volumes is an essential step for the quantitative diagnosis and treatment of many vascular diseases. 3D vessel segmentation is being actively investigated in existing works, mostly in deep learning approaches. However, training 3D deep networks requires large amounts of manual 3D annotations from experts, which are laborious to obtain. This is especially the case for 3D vessel segmentation, as vessels are sparse yet spread out over many slices and disconnected when visualized in 2D slices. In this work, we propose a novel method to segment the 3D peripancreatic arteries solely from one annotated 2D projection per training image with depth supervision. We perform extensive experiments on the segmentation of peripancreatic arteries on 3D contrast-enhanced CT images and demonstrate how well we capture the rich depth information from 2D projections. We demonstrate that by annotating a single, randomly chosen projection for each training sample, we obtain comparable performance to annotating multiple 2D projections, thereby reducing the annotation effort. Furthermore, by mapping the 2D labels to the 3D space using depth information and incorporating this into training, we almost close the performance gap between 3D supervision and 2D supervision. Our code is available at: https://github.com/alinafdima/3Dseg-mip-depth.

Bailiang Jian, Mohammad Farid Azampour, Francesca De Benetti et al.

CT and MRI are two of the most informative modalities in spinal diagnostics and treatment planning. CT is useful when analysing bony structures, while MRI gives information about the soft tissue. Thus, fusing the information of both modalities can be very beneficial. Registration is the first step for this fusion. While the soft tissues around the vertebra are deformable, each vertebral body is constrained to move rigidly. We propose a weakly-supervised deep learning framework that preserves the rigidity and the volume of each vertebra while maximizing the accuracy of the registration. To achieve this goal, we introduce anatomy-aware losses for training the network. We specifically design these losses to depend only on the CT label maps since automatic vertebra segmentation in CT gives more accurate results contrary to MRI. We evaluate our method on an in-house dataset of 167 patients. Our results show that adding the anatomy-aware losses increases the plausibility of the inferred transformation while keeping the accuracy untouched.

Diana Waldmannstetter, Benedikt Wiestler, Julian Schwarting et al.

Even though simultaneous optimization of similarity metrics is a standard procedure in the field of semantic segmentation, surprisingly, this is much less established for image registration. To help closing this gap in the literature, we investigate in a complex multi-modal 3D setting whether simultaneous optimization of registration metrics, here implemented by means of primitive summation, can benefit image registration. We evaluate two challenging datasets containing collections of pre- to post-operative and pre- to intra-operative MR images of glioma. Employing the proposed optimization, we demonstrate improved registration accuracy in terms of TRE on expert neuroradiologists' landmark annotations.

Ezequiel de la Rosa, Ruisheng Su, Mauricio Reyes et al.

Accurate estimation of brain infarction (i.e., irreversibly damaged tissue) is critical for guiding treatment decisions in acute ischemic stroke. Reliable infarct prediction informs key clinical interventions, including the need for patient transfer to comprehensive stroke centers, the potential benefit of additional reperfusion attempts during mechanical thrombectomy, decisions regarding secondary neuroprotective treatments, and ultimately, prognosis of clinical outcomes. This work introduces the Ischemic Stroke Lesion Segmentation (ISLES) 2024 challenge, which focuses on the prediction of final infarct volumes from pre-interventional acute stroke imaging and clinical data. ISLES24 provides a comprehensive, multimodal setting where participants can leverage all clinically and practically available data, including full acute CT imaging, sub-acute follow-up MRI, and structured clinical information, across a train set of 150 cases. On the hidden test set of 98 cases, the top-performing model, a multimodal nnU-Net-based architecture, achieved a Dice score of 0.285 (+/- 0.213) and an absolute volume difference of 21.2 (+/- 37.2) mL, underlining the significant challenges posed by this task and the need for further advances in multimodal learning. This work makes two primary contributions: first, we establish a standardized, clinically realistic benchmark for post-treatment infarct prediction, enabling systematic evaluation of multimodal algorithmic strategies on a longitudinal stroke dataset; second, we analyze current methodological limitations and outline key research directions to guide the development of next-generation infarct prediction models.

Matthias Keicher, Matan Atad, David Schinz et al.

Vertebral fractures are a consequence of osteoporosis, with significant health implications for affected patients. Unfortunately, grading their severity using CT exams is hard and subjective, motivating automated grading methods. However, current approaches are hindered by imbalance and scarcity of data and a lack of interpretability. To address these challenges, this paper proposes a novel approach that leverages unlabelled data to train a generative Diffusion Autoencoder (DAE) model as an unsupervised feature extractor. We model fracture grading as a continuous regression, which is more reflective of the smooth progression of fractures. Specifically, we use a binary, supervised fracture classifier to construct a hyperplane in the DAE's latent space. We then regress the severity of the fracture as a function of the distance to this hyperplane, calibrating the results to the Genant scale. Importantly, the generative nature of our method allows us to visualize different grades of a given vertebra, providing interpretability and insight into the features that contribute to automated grading.

Matan Atad, Alexander W. Marka, Lisa Steinhelfer et al.

Accurate segmentation of vertebral metastasis in CT is clinically important yet difficult to scale, as voxel-level annotations are scarce and both lytic and blastic lesions often resemble benign degenerative changes. We introduce a weakly supervised method trained solely on vertebra-level healthy/malignant labels, without any lesion masks. The method combines a Diffusion Autoencoder (DAE) that produces a classifier-guided healthy edit of each vertebra with pixel-wise difference maps that propose candidate lesion regions. To determine which regions truly reflect malignancy, we introduce Hide-and-Seek Attribution: each candidate is revealed in turn while all others are hidden, the edited image is projected back to the data manifold by the DAE, and a latent-space classifier quantifies the isolated malignant contribution of that component. High-scoring regions form the final lytic or blastic segmentation. On held-out radiologist annotations, we achieve strong blastic/lytic performance despite no mask supervision (F1: 0.91/0.85; Dice: 0.87/0.78), exceeding baselines (F1: 0.79/0.67; Dice: 0.74/0.55). These results show that vertebra-level labels can be transformed into reliable lesion masks, demonstrating that generative editing combined with selective occlusion supports accurate weakly supervised segmentation in CT.

Diana Waldmannstetter, Ivan Ezhov, Benedikt Wiestler et al.

Accurate image registration is pivotal in biomedical image analysis, where selecting suitable registration algorithms demands careful consideration. While numerous algorithms are available, the evaluation metrics to assess their performance have remained relatively static. This study addresses this challenge by introducing a novel evaluation metric termed Landmark Hit Rate (HitR), which focuses on the clinical relevance of image registration accuracy. Unlike traditional metrics such as Target Registration Error, which emphasize subresolution differences, HitR considers whether registration algorithms successfully position landmarks within defined confidence zones. This paradigm shift acknowledges the inherent annotation noise in medical images, allowing for more meaningful assessments. To equip HitR with label-noise-awareness, we propose defining these confidence zones based on an Inter-rater Variance analysis. Consequently, hit rate curves are computed for varying landmark zone sizes, enabling performance measurement for a task-specific level of accuracy. Our approach offers a more realistic and meaningful assessment of image registration algorithms, reflecting their suitability for clinical and biomedical applications.

Matan Atad, Gabriel Gruber, Marx Ribeiro et al.

Accurate calibration of finite element (FE) models is essential across various biomechanical applications, including human intervertebral discs (IVDs), to ensure their reliability and use in diagnosing and planning treatments. However, traditional calibration methods are computationally intensive, requiring iterative, derivative-free optimization algorithms that often take days to converge. This study addresses these challenges by introducing a novel, efficient, and effective calibration method demonstrated on a human L4-L5 IVD FE model as a case study using a neural network (NN) surrogate. The NN surrogate predicts simulation outcomes with high accuracy, outperforming other machine learning models, and significantly reduces the computational cost associated with traditional FE simulations. Next, a Projected Gradient Descent (PGD) approach guided by gradients of the NN surrogate is proposed to efficiently calibrate FE models. Our method explicitly enforces feasibility with a projection step, thus maintaining material bounds throughout the optimization process. The proposed method is evaluated against SOTA Genetic Algorithm and inverse model baselines on synthetic and in vitro experimental datasets. Our approach demonstrates superior performance on synthetic data, achieving an MAE of 0.06 compared to the baselines' MAE of 0.18 and 0.54, respectively. On experimental specimens, our method outperforms the baseline in 5 out of 6 cases. While our approach requires initial dataset generation and surrogate training, these steps are performed only once, and the actual calibration takes under three seconds. In contrast, traditional calibration time scales linearly with the number of specimens, taking up to 8 days in the worst-case. Such efficiency paves the way for applying more complex FE models, potentially extending beyond IVDs, and enabling accurate patient-specific simulations.

Nathan Molinier, Hendrik Möller, Thomas Dagonneau et al.

Deep learning-based medical image segmentation is increasingly used to support clinical diagnosis and develop new treatment strategies. However, model performance remains limited by the scarcity of high-quality annotated data and insufficient generalization across imaging protocols. This limitation is particularly evident in MRI and CT, where models are typically trained on a single acquisition sequence and exhibit reduced robustness when applied to unseen sequences or contrasts. Although data augmentation is widely used to improve general robustness on medical images, its impact on cross-modality generalization has not been quantitatively explored. In this work, we study a targeted set of data augmentation techniques designed to improve cross-modality transfer. We train three spine segmentation models, each on a single-modality/sequence dataset, and evaluate them across seven out-of-distribution datasets (spanning CT and MRI), reflecting a realistic single-sequence training and multi-sequence/contrast/modality deployment scenario. Our results demonstrate substantial performance gains on unseen domains (average Dice gain of 155 %) while preserving in-domain accuracy (average Dice decrease of 0.008 %), including effective transfer between CT and MRI. To mitigate the computational cost typically associated with strong data augmentation, we implement GPU-optimized augmentations that maintain, and even improve, training efficiency by approximately 10 %. We release our approach as an open-source toolbox, enabling seamless integration into commonly used frameworks such as nnUNet and MONAI. These augmentations significantly enhance robustness to heterogeneous clinical imaging scenarios without compromising training speed.

Hendrik Möller, Robert Graf, Joachim Schmitt et al.

Purpose. To present SPINEPS, an open-source deep learning approach for semantic and instance segmentation of 14 spinal structures (ten vertebra substructures, intervertebral discs, spinal cord, spinal canal, and sacrum) in whole body T2w MRI. Methods. During this HIPPA-compliant, retrospective study, we utilized the public SPIDER dataset (218 subjects, 63% female) and a subset of the German National Cohort (1423 subjects, mean age 53, 49% female) for training and evaluation. We combined CT and T2w segmentations to train models that segment 14 spinal structures in T2w sagittal scans both semantically and instance-wise. Performance evaluation metrics included Dice similarity coefficient, average symmetrical surface distance, panoptic quality, segmentation quality, and recognition quality. Statistical significance was assessed using the Wilcoxon signed-rank test. An in-house dataset was used to qualitatively evaluate out-of-distribution samples. Results. On the public dataset, our approach outperformed the baseline (instance-wise vertebra dice score 0.929 vs. 0.907, p-value<0.001). Training on auto-generated annotations and evaluating on manually corrected test data from the GNC yielded global dice scores of 0.900 for vertebrae, 0.960 for intervertebral discs, and 0.947 for the spinal canal. Incorporating the SPIDER dataset during training increased these scores to 0.920, 0.967, 0.958, respectively. Conclusions. The proposed segmentation approach offers robust segmentation of 14 spinal structures in T2w sagittal images, including the spinal cord, spinal canal, intervertebral discs, endplate, sacrum, and vertebrae. The approach yields both a semantic and instance mask as output, thus being easy to utilize. This marks the first publicly available algorithm for whole spine segmentation in sagittal T2w MR imaging.

Ezequiel de la Rosa, Mauricio Reyes, Sook-Lei Liew et al.

Diffusion-weighted MRI (DWI) is essential for stroke diagnosis, treatment decisions, and prognosis. However, image and disease variability hinder the development of generalizable AI algorithms with clinical value. We address this gap by presenting a novel ensemble algorithm derived from the 2022 Ischemic Stroke Lesion Segmentation (ISLES) challenge. ISLES'22 provided 400 patient scans with ischemic stroke from various medical centers, facilitating the development of a wide range of cutting-edge segmentation algorithms by the research community. Through collaboration with leading teams, we combined top-performing algorithms into an ensemble model that overcomes the limitations of individual solutions. Our ensemble model achieved superior ischemic lesion detection and segmentation accuracy on our internal test set compared to individual algorithms. This accuracy generalized well across diverse image and disease variables. Furthermore, the model excelled in extracting clinical biomarkers. Notably, in a Turing-like test, neuroradiologists consistently preferred the algorithm's segmentations over manual expert efforts, highlighting increased comprehensiveness and precision. Validation using a real-world external dataset (N=1686) confirmed the model's generalizability. The algorithm's outputs also demonstrated strong correlations with clinical scores (admission NIHSS and 90-day mRS) on par with or exceeding expert-derived results, underlining its clinical relevance. This study offers two key findings. First, we present an ensemble algorithm (https://github.com/Tabrisrei/ISLES22_Ensemble) that detects and segments ischemic stroke lesions on DWI across diverse scenarios on par with expert (neuro)radiologists. Second, we show the potential for biomedical challenge outputs to extend beyond the challenge's initial objectives, demonstrating their real-world clinical applicability.

Florian Kofler, Hendrik Möller, Josef A. Buchner et al.

This paper introduces panoptica, a versatile and performance-optimized package designed for computing instance-wise segmentation quality metrics from 2D and 3D segmentation maps. panoptica addresses the limitations of existing metrics and provides a modular framework that complements the original intersection over union-based panoptic quality with other metrics, such as the distance metric Average Symmetric Surface Distance. The package is open-source, implemented in Python, and accompanied by comprehensive documentation and tutorials. panoptica employs a three-step metrics computation process to cover diverse use cases. The efficacy of panoptica is demonstrated on various real-world biomedical datasets, where an instance-wise evaluation is instrumental for an accurate representation of the underlying clinical task. Overall, we envision panoptica as a valuable tool facilitating in-depth evaluation of segmentation methods.

Jun Ma, Feifei Li, Sumin Kim et al.

Promptable segmentation foundation models have emerged as a transformative approach to addressing the diverse needs in medical images, but most existing models require expensive computing, posing a big barrier to their adoption in clinical practice. In this work, we organized the first international competition dedicated to promptable medical image segmentation, featuring a large-scale dataset spanning nine common imaging modalities from over 20 different institutions. The top teams developed lightweight segmentation foundation models and implemented an efficient inference pipeline that substantially reduced computational requirements while maintaining state-of-the-art segmentation accuracy. Moreover, the post-challenge phase advanced the algorithms through the design of performance booster and reproducibility tasks, resulting in improved algorithms and validated reproducibility of the winning solution. Furthermore, the best-performing algorithms have been incorporated into the open-source software with a user-friendly interface to facilitate clinical adoption. The data and code are publicly available to foster the further development of medical image segmentation foundation models and pave the way for impactful real-world applications.

Murong Xu, Tamaz Amiranashvili, Fernando Navarro et al.

Accurate delineation of anatomical structures in volumetric CT scans is crucial for diagnosis and treatment planning. While AI has advanced automated segmentation, current approaches typically target individual structures, creating a fragmented landscape of incompatible models with varying performance and disparate evaluation protocols. Foundational segmentation models address these limitations by providing a holistic anatomical view through a single model. Yet, robust clinical deployment demands comprehensive training data, which is lacking in existing whole-body approaches, both in terms of data heterogeneity and, more importantly, anatomical coverage. In this work, rather than pursuing incremental optimizations in model architecture, we present CADS, an open-source framework that prioritizes the systematic integration, standardization, and labeling of heterogeneous data sources for whole-body CT segmentation. At its core is a large-scale dataset of 22,022 CT volumes with complete annotations for 167 anatomical structures, representing a significant advancement in both scale and coverage, with 18 times more scans than existing collections and 60% more distinct anatomical targets. Building on this diverse dataset, we develop the CADS-model using established architectures for accessible and automated full-body CT segmentation. Through comprehensive evaluation across 18 public datasets and an independent real-world hospital cohort, we demonstrate advantages over SoTA approaches. Notably, thorough testing of the model's performance in segmentation tasks from radiation oncology validates its direct utility for clinical interventions. By making our large-scale dataset, our segmentation models, and our clinical software tool publicly available, we aim to advance robust AI solutions in radiology and make comprehensive anatomical analysis accessible to clinicians and researchers alike.

Anjany Sekuboyina, Malek E. Husseini, Amirhossein Bayat et al.

Vertebral labelling and segmentation are two fundamental tasks in an automated spine processing pipeline. Reliable and accurate processing of spine images is expected to benefit clinical decision-support systems for diagnosis, surgery planning, and population-based analysis on spine and bone health. However, designing automated algorithms for spine processing is challenging predominantly due to considerable variations in anatomy and acquisition protocols and due to a severe shortage of publicly available data. Addressing these limitations, the Large Scale Vertebrae Segmentation Challenge (VerSe) was organised in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2019 and 2020, with a call for algorithms towards labelling and segmentation of vertebrae. Two datasets containing a total of 374 multi-detector CT scans from 355 patients were prepared and 4505 vertebrae have individually been annotated at voxel-level by a human-machine hybrid algorithm (https://osf.io/nqjyw/, https://osf.io/t98fz/). A total of 25 algorithms were benchmarked on these datasets. In this work, we present the the results of this evaluation and further investigate the performance-variation at vertebra-level, scan-level, and at different fields-of-view. We also evaluate the generalisability of the approaches to an implicit domain shift in data by evaluating the top performing algorithms of one challenge iteration on data from the other iteration. The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations. The content and code concerning VerSe can be accessed at: https://github.com/anjany/verse.

Kaiyuan Yang, Fabio Musio, Yihui Ma et al.

The Circle of Willis (CoW) is an important network of arteries connecting major circulations of the brain. Its vascular architecture is believed to affect the risk, severity, and clinical outcome of serious neurovascular diseases. However, characterizing the highly variable CoW anatomy is still a manual and time-consuming expert task. The CoW is usually imaged by two non-invasive angiographic imaging modalities, magnetic resonance angiography (MRA) and computed tomography angiography (CTA), but there exist limited datasets with annotations on CoW anatomy, especially for CTA. Therefore, we organized the TopCoW challenge with the release of an annotated CoW dataset. The TopCoW dataset is the first public dataset with voxel-level annotations for 13 CoW vessel components, enabled by virtual reality technology. It is also the first large dataset using 200 pairs of MRA and CTA from the same patients. As part of the benchmark, we invited submissions worldwide and attracted over 250 registered participants from six continents. The submissions were evaluated on both internal and external test datasets of 226 scans from over five centers. The top performing teams achieved over 90% Dice scores at segmenting the CoW components, over 80% F1 scores at detecting key CoW components, and over 70% balanced accuracy at classifying CoW variants for nearly all test sets. The best algorithms also showed clinical potential in classifying fetal-type posterior cerebral artery and locating aneurysms with CoW anatomy. TopCoW demonstrated the utility and versatility of CoW segmentation algorithms for a wide range of downstream clinical applications with explainability. The annotated datasets and best performing algorithms have been released as public Zenodo records to foster further methodological development and clinical tool building.

Hendrik Möller, Hanna Schön, Alina Dima et al.

Thoracolumbar stump ribs are one of the essential indicators of thoracolumbar transitional vertebrae or enumeration anomalies. While some studies manually assess these anomalies and describe the ribs qualitatively, this study aims to automate thoracolumbar stump rib detection and analyze their morphology quantitatively. To this end, we train a high-resolution deep-learning model for rib segmentation and show significant improvements compared to existing models (Dice score 0.997 vs. 0.779, p-value < 0.01). In addition, we use an iterative algorithm and piece-wise linear interpolation to assess the length of the ribs, showing a success rate of 98.2%. When analyzing morphological features, we show that stump ribs articulate more posteriorly at the vertebrae (-19.2 +- 3.8 vs -13.8 +- 2.5, p-value < 0.01), are thinner (260.6 +- 103.4 vs. 563.6 +- 127.1, p-value < 0.01), and are oriented more downwards and sideways within the first centimeters in contrast to full-length ribs. We show that with partially visible ribs, these features can achieve an F1-score of 0.84 in differentiating stump ribs from regular ones. We publish the model weights and masks for public use.

Robert Graf, Tanja Lerchl, Kati Nispel et al.

Digital twins offer a powerful framework for subject-specific simulation and clinical decision support, yet their development often hinges on accurate, individualized anatomical modeling. In this work, we present a rule-based approach for subpixel-accurate key-point extraction from MRI, adapted from prior CT-based methods. Our approach incorporates robust image alignment and vertebra-specific orientation estimation to generate anatomically meaningful landmarks that serve as boundary conditions and force application points, like muscle and ligament insertions in biomechanical models. These models enable the simulation of spinal mechanics considering the subject's individual anatomy, and thus support the development of tailored approaches in clinical diagnostics and treatment planning. By leveraging MR imaging, our method is radiation-free and well-suited for large-scale studies and use in underrepresented populations. This work contributes to the digital twin ecosystem by bridging the gap between precise medical image analysis with biomechanical simulation, and aligns with key themes in personalized modeling for healthcare.

Paul Engstler, Matthias Keicher, David Schinz et al.

Do black-box neural network models learn clinically relevant features for fracture diagnosis? The answer not only establishes reliability quenches scientific curiosity but also leads to explainable and verbose findings that can assist the radiologists in the final and increase trust. This work identifies the concepts networks use for vertebral fracture diagnosis in CT images. This is achieved by associating concepts to neurons highly correlated with a specific diagnosis in the dataset. The concepts are either associated with neurons by radiologists pre-hoc or are visualized during a specific prediction and left for the user's interpretation. We evaluate which concepts lead to correct diagnosis and which concepts lead to false positives. The proposed frameworks and analysis pave the way for reliable and explainable vertebral fracture diagnosis.

Ezequiel de la Rosa, David Robben, Diana M. Sima et al.

Perfusion imaging is the current gold standard for acute ischemic stroke analysis. It allows quantification of the salvageable and non-salvageable tissue regions (penumbra and core areas respectively). In clinical settings, the singular value decomposition (SVD) deconvolution is one of the most accepted and used approaches for generating interpretable and physically meaningful maps. Though this method has been widely validated in experimental and clinical settings, it might produce suboptimal results because the chosen inputs to the model cannot guarantee optimal performance. For the most critical input, the arterial input function (AIF), it is still controversial how and where it should be chosen even though the method is very sensitive to this input. In this work we propose an AIF selection approach that is optimized for maximal core lesion segmentation performance. The AIF is regressed by a neural network optimized through a differentiable SVD deconvolution, aiming to maximize core lesion segmentation agreement with ground truth data. To our knowledge, this is the first work exploiting a differentiable deconvolution model with neural networks. We show that our approach is able to generate AIFs without any manual annotation, and hence avoiding manual rater's influences. The method achieves manual expert performance in the ISLES18 dataset. We conclude that the methodology opens new possibilities for improving perfusion imaging quantification with deep neural networks.

Christina Bukas, Bailiang Jian, Luis F. Rodriguez Venegas et al.

Symptomatic spinal vertebral compression fractures (VCFs) often require osteoplasty treatment. A cement-like material is injected into the bone to stabilize the fracture, restore the vertebral body height and alleviate pain. Leakage is a common complication and may occur due to too much cement being injected. In this work, we propose an automated patient-specific framework that can allow physicians to calculate an upper bound of cement for the injection and estimate the optimal outcome of osteoplasty. The framework uses the patient CT scan and the fractured vertebra label to build a virtual healthy spine using a high-level approach. Firstly, the fractured spine is segmented with a three-step Convolution Neural Network (CNN) architecture. Next, a per-vertebra rigid registration to a healthy spine atlas restores its curvature. Finally, a GAN-based inpainting approach replaces the fractured vertebra with an estimation of its original shape. Based on this outcome, we then estimate the maximum amount of bone cement for injection. We evaluate our framework by comparing the virtual vertebrae volumes of ten patients to their healthy equivalent and report an average error of 3.88$\pm$7.63\%. The presented pipeline offers a first approach to a personalized automatic high-level framework for planning osteoplasty procedures.

Hans Liebl, David Schinz, Anjany Sekuboyina et al.

With the advent of deep learning algorithms, fully automated radiological image analysis is within reach. In spine imaging, several atlas- and shape-based as well as deep learning segmentation algorithms have been proposed, allowing for subsequent automated analysis of morphology and pathology. The first Large Scale Vertebrae Segmentation Challenge (VerSe 2019) showed that these perform well on normal anatomy, but fail in variants not frequently present in the training dataset. Building on that experience, we report on the largely increased VerSe 2020 dataset and results from the second iteration of the VerSe challenge (MICCAI 2020, Lima, Peru). VerSe 2020 comprises annotated spine computed tomography (CT) images from 300 subjects with 4142 fully visualized and annotated vertebrae, collected across multiple centres from four different scanner manufacturers, enriched with cases that exhibit anatomical variants such as enumeration abnormalities (n=77) and transitional vertebrae (n=161). Metadata includes vertebral labelling information, voxel-level segmentation masks obtained with a human-machine hybrid algorithm and anatomical ratings, to enable the development and benchmarking of robust and accurate segmentation algorithms.

Ezequiel de la Rosa, Diana M. Sima, Bjoern Menze et al.

Perfusion imaging is crucial in acute ischemic stroke for quantifying the salvageable penumbra and irreversibly damaged core lesions. As such, it helps clinicians to decide on the optimal reperfusion treatment. In perfusion CT imaging, deconvolution methods are used to obtain clinically interpretable perfusion parameters that allow identifying brain tissue abnormalities. Deconvolution methods require the selection of two reference vascular functions as inputs to the model: the arterial input function (AIF) and the venous output function, with the AIF as the most critical model input. When manually performed, the vascular function selection is time demanding, suffers from poor reproducibility and is subject to the professionals' experience. This leads to potentially unreliable quantification of the penumbra and core lesions and, hence, might harm the treatment decision process. In this work we automatize the perfusion analysis with AIFNet, a fully automatic and end-to-end trainable deep learning approach for estimating the vascular functions. Unlike previous methods using clustering or segmentation techniques to select vascular voxels, AIFNet is directly optimized at the vascular function estimation, which allows to better recognise the time-curve profiles. Validation on the public ISLES18 stroke database shows that AIFNet reaches inter-rater performance for the vascular function estimation and, subsequently, for the parameter maps and core lesion quantification obtained through deconvolution. We conclude that AIFNet has potential for clinical transfer and could be incorporated in perfusion deconvolution software.

Hanwool Park, Amirhossein Bayat, Mohammad Sabokrou et al.

This paper presents a novel yet efficient defense framework for segmentation models against adversarial attacks in medical imaging. In contrary to the defense methods against adversarial attacks for classification models which widely are investigated, such defense methods for segmentation models has been less explored. Our proposed method can be used for any deep learning models without revising the target deep learning models, as well as can be independent of adversarial attacks. Our framework consists of a frequency domain converter, a detector, and a reformer. The frequency domain converter helps the detector detects adversarial examples by using a frame domain of an image. The reformer helps target models to predict more precisely. We have experiments to empirically show that our proposed method has a better performance compared to the existing defense method.

Amirhossein Bayat, Suprosanna Shit, Adrian Kilian et al.

Designing of a cranial implant needs a 3D understanding of the complete skull shape. Thus, taking a 2D approach is sub-optimal, since a 2D model lacks a holistic 3D view of both the defective and healthy skulls. Further, loading the whole 3D skull shapes at its original image resolution is not feasible in commonly available GPUs. To mitigate these issues, we propose a fully convolutional network composed of two subnetworks. The first subnetwork is designed to complete the shape of the downsampled defective skull. The second subnetwork upsamples the reconstructed shape slice-wise. We train the 3D and 2D networks together end-to-end, with a hierarchical loss function. Our proposed solution accurately predicts a high-resolution 3D implant in the challenge test case in terms of dice-score and the Hausdorff distance.

Malek Husseini, Anjany Sekuboyina, Maximilian Loeffler et al.

Osteoporotic vertebral fractures have a severe impact on patients' overall well-being but are severely under-diagnosed. These fractures present themselves at various levels of severity measured using the Genant's grading scale. Insufficient annotated datasets, severe data-imbalance, and minor difference in appearances between fractured and healthy vertebrae make naive classification approaches result in poor discriminatory performance. Addressing this, we propose a representation learning-inspired approach for automated vertebral fracture detection, aimed at learning latent representations efficient for fracture detection. Building on state-of-art metric losses, we present a novel Grading Loss for learning representations that respect Genant's fracture grading scheme. On a publicly available spine dataset, the proposed loss function achieves a fracture detection F1 score of 81.5%, a 10% increase over a naive classification baseline.

Amirhossein Bayat, Anjany Sekuboyina, Johannes C. Paetzold et al.

The treatment of degenerative spinal disorders requires an understanding of the individual spinal anatomy and curvature in 3D. An upright spinal pose (i.e. standing) under natural weight bearing is crucial for such bio-mechanical analysis. 3D volumetric imaging modalities (e.g. CT and MRI) are performed in patients lying down. On the other hand, radiographs are captured in an upright pose, but result in 2D projections. This work aims to integrate the two realms, i.e. it combines the upright spinal curvature from radiographs with the 3D vertebral shape from CT imaging for synthesizing an upright 3D model of spine, loaded naturally. Specifically, we propose a novel neural network architecture working vertebra-wise, termed \emph{TransVert}, which takes orthogonal 2D radiographs and infers the spine's 3D posture. We validate our architecture on digitally reconstructed radiographs, achieving a 3D reconstruction Dice of $95.52\%$, indicating an almost perfect 2D-to-3D domain translation. Deploying our model on clinical radiographs, we successfully synthesise full-3D, upright, patient-specific spine models for the first time.

Anjany Sekuboyina, Markus Rempfler, Alexander Valentinitsch et al.

We propose an auto-encoding network architecture for point clouds (PC) capable of extracting shape signatures without supervision. Building on this, we (i) design a loss function capable of modelling data variance on PCs which are unstructured, and (ii) regularise the latent space as in a variational auto-encoder, both of which increase the auto-encoders' descriptive capacity while making them probabilistic. Evaluating the reconstruction quality of our architectures, we employ them for detecting vertebral fractures without any supervision. By learning to efficiently reconstruct only healthy vertebrae, fractures are detected as anomalous reconstructions. Evaluating on a dataset containing $\sim$1500 vertebrae, we achieve area-under-ROC curve of $>$75%, without using intensity-based features.

Hongwei Li, Johannes C. Paetzold, Anjany Sekuboyina et al.

Synthesizing MR imaging sequences is highly relevant in clinical practice, as single sequences are often missing or are of poor quality (e.g. due to motion). Naturally, the idea arises that a target modality would benefit from multi-modal input, as proprietary information of individual modalities can be synergistic. However, existing methods fail to scale up to multiple non-aligned imaging modalities, facing common drawbacks of complex imaging sequences. We propose a novel, scalable and multi-modal approach called DiamondGAN. Our model is capable of performing exible non-aligned cross-modality synthesis and data infill, when given multiple modalities or any of their arbitrary subsets, learning structured information in an end-to-end fashion. We synthesize two MRI sequences with clinical relevance (i.e., double inversion recovery (DIR) and contrast-enhanced T1 (T1-c)), reconstructed from three common sequences. In addition, we perform a multi-rater visual evaluation experiment and find that trained radiologists are unable to distinguish synthetic DIR images from real ones.

Anjany Sekuboyina, Markus Rempfler, Alexander Valentinitsch et al.

Purpose: To use and test a labelling algorithm that operates on two-dimensional (2D) reformations, rather than three-dimensional (3D) data to locate and identify vertebrae. Methods: We improved the Btrfly Net (described by Sekuboyina et al) that works on sagittal and coronal maximum intensity projections (MIP) and augmented it with two additional components: spine-localization and adversarial a priori-learning. Furthermore, we explored two variants of adversarial training schemes that incorporated the anatomical a priori knowledge into the Btrfly Net. We investigated the superiority of the proposed approach for labelling vertebrae on three datasets: a public benchmarking dataset of 302 CT scans and two in-house datasets with a total of 238 CT scans. We employed Wilcoxon signed-rank test to compute the statistical significance of the improvement in performance observed due to various architectural components in our approach. Results: On the public dataset, our approach using the described Btrfly(pe-eb) network performed on par with current state-of-the-art methods achieving a statistically significant (p < .001) vertebrae identification rate of 88.5+/-0.2 % and localization distances of less than 7-mm. On the in-house datasets that had a higher inter-scan data variability, we obtained an identification rate of 85.1+/-1.2%. Conclusion: An identification performance comparable to existing 3D approaches was achieved when labelling vertebrae on 2D MIPs. The performance was further improved using the proposed adversarial training regime that effectively enforced local spine a priori knowledge during training. Lastly, spine-localization increased the generalizability of our approach by homogenizing the content in the MIPs.

Ezequiel de la Rosa, Diana M. Sima, Thijs Vande Vyvere et al.

Computer Tomography (CT) is the gold standard technique for brain damage evaluation after acute Traumatic Brain Injury (TBI). It allows identification of most lesion types and determines the need of surgical or alternative therapeutic procedures. However, the traditional approach for lesion classification is restricted to visual image inspection. In this work, we characterize and predict TBI lesions by using CT-derived radiomics descriptors. Relevant shape, intensity and texture biomarkers characterizing the different lesions are isolated and a lesion predictive model is built by using Partial Least Squares. On a dataset containing 155 scans (105 train, 50 test) the methodology achieved 89.7 % accuracy over the unseen data. When a model was build using only texture features, a 88.2 % accuracy was obtained. Our results suggest that selected radiomics descriptors could play a key role in brain injury prediction. Besides, the proposed methodology is close to reproduce radiologists decision making. These results open new possibilities for radiomics-inspired brain lesion detection, segmentation and prediction.

Anjany Sekuboyina, Markus Rempfler, Jan Kukačka et al.

Robust localisation and identification of vertebrae is essential for automated spine analysis. The contribution of this work to the task is two-fold: (1) Inspired by the human expert, we hypothesise that a sagittal and coronal reformation of the spine contain sufficient information for labelling the vertebrae. Thereby, we propose a butterfly-shaped network architecture (termed Btrfly Net) that efficiently combines the information across reformations. (2) Underpinning the Btrfly net, we present an energy-based adversarial training regime that encodes local spine structure as an anatomical prior into the network, thereby enabling it to achieve state-of-art performance in all standard metrics on a benchmark dataset of 302 scans without any post-processing during inference.

Anjany Sekuboyina, Alexander Valentinitsch, Jan S. Kirschke et al.

Multi-class segmentation of vertebrae is a non-trivial task mainly due to the high correlation in the appearance of adjacent vertebrae. Hence, such a task calls for the consideration of both global and local context. Based on this motivation, we propose a two-staged approach that, given a computed tomography dataset of the spine, segments the five lumbar vertebrae and simultaneously labels them. The first stage employs a multi-layered perceptron performing non-linear regression for locating the lumbar region using the global context. The second stage, comprised of a fully-convolutional deep network, exploits the local context in the localised lumbar region to segment and label the lumbar vertebrae in one go. Aided with practical data augmentation for training, our approach is highly generalisable, capable of successfully segmenting both healthy and abnormal vertebrae (fractured and scoliotic spines). We consistently achieve an average Dice coefficient of over 90 percent on a publicly available dataset of the xVertSeg segmentation challenge of MICCAI 2016. This is particularly noteworthy because the xVertSeg dataset is beset with severe deformities in the form of vertebral fractures and scoliosis.