Christian Wachinger

Jianning Li, Zongwei Zhou, Jiancheng Yang et al.

Prior to the deep learning era, shape was commonly used to describe the objects. Nowadays, state-of-the-art (SOTA) algorithms in medical imaging are predominantly diverging from computer vision, where voxel grids, meshes, point clouds, and implicit surface models are used. This is seen from numerous shape-related publications in premier vision conferences as well as the growing popularity of ShapeNet (about 51,300 models) and Princeton ModelNet (127,915 models). For the medical domain, we present a large collection of anatomical shapes (e.g., bones, organs, vessels) and 3D models of surgical instrument, called MedShapeNet, created to facilitate the translation of data-driven vision algorithms to medical applications and to adapt SOTA vision algorithms to medical problems. As a unique feature, we directly model the majority of shapes on the imaging data of real patients. As of today, MedShapeNet includes 23 dataset with more than 100,000 shapes that are paired with annotations (ground truth). Our data is freely accessible via a web interface and a Python application programming interface (API) and can be used for discriminative, reconstructive, and variational benchmarks as well as various applications in virtual, augmented, or mixed reality, and 3D printing. Exemplary, we present use cases in the fields of classification of brain tumors, facial and skull reconstructions, multi-class anatomy completion, education, and 3D printing. In future, we will extend the data and improve the interfaces. The project pages are: https://medshapenet.ikim.nrw/ and https://github.com/Jianningli/medshapenet-feedback

Tom Nuno Wolf, Sebastian Pölsterl, Christian Wachinger

Alzheimer's disease (AD) has a complex and multifactorial etiology, which requires integrating information about neuroanatomy, genetics, and cerebrospinal fluid biomarkers for accurate diagnosis. Hence, recent deep learning approaches combined image and tabular information to improve diagnostic performance. However, the black-box nature of such neural networks is still a barrier for clinical applications, in which understanding the decision of a heterogeneous model is integral. We propose PANIC, a prototypical additive neural network for interpretable AD classification that integrates 3D image and tabular data. It is interpretable by design and, thus, avoids the need for post-hoc explanations that try to approximate the decision of a network. Our results demonstrate that PANIC achieves state-of-the-art performance in AD classification, while directly providing local and global explanations. Finally, we show that PANIC extracts biologically meaningful signatures of AD, and satisfies a set of desirable desiderata for trustworthy machine learning. Our implementation is available at https://github.com/ai-med/PANIC .

Bailiang Jian, Jiazhen Pan, Morteza Ghahremani et al.

Our findings indicate that adopting "advanced" computational elements fails to significantly improve registration accuracy. Instead, well-established registration-specific designs offer fair improvements, enhancing results by a marginal 1.5\% over the baseline. Our findings emphasize the importance of rigorous, unbiased evaluation and contribution disentanglement of all low- and high-level registration components, rather than simply following the computer vision trends with "more advanced" computational blocks. We advocate for simpler yet effective solutions and novel evaluation metrics that go beyond conventional registration accuracy, warranting further research across diverse organs and modalities. The code is available at \url{https://github.com/BailiangJ/rethink-reg}.

Tom Maye-Lasserre, Yitong Li, Bailiang Jian et al.

Modern 3D medical vision-language models (VLMs) can generate fluent radiology-style text while exhibit critically low pathology detection and output diversity, collapsing to generic templates that under-report rare yet critical findings. We identify this failure mode as Template Collapse. This failure stems from the unique constraints of 3D medical imaging, e.g., limited data, severe label imbalance, and weak signals from volumetric encoders. Under these constraints, text-generation objectives encourage shortcut learning and fluent but weakly grounded reports. We systematically diagnose the Template Collapse through clinical fidelity, output diversity, normal-template bias, and rare-finding survival. To mitigate it, we propose CLarGen, a decoupled framework that separates what to say (clinical detection) from how to say it (language synthesis). CLarGen uses (i) a Latent Query Transformer for multi-label pathology detection, (ii) pathology-guided retrieval for clinically matched exemplars, and (iii) a medical language model to synthesize the final report from detected findings and retrieved context. Across state-of-the-art 3D CT report generation baselines, CLarGen mitigates Template Collapse and substantially improves clinical accuracy (macro-F1 0.487 vs. 0.189; CRG 0.472 vs. 0.368) while maintaining fluent reporting. Our results suggest that explicit, measurable clinical grounding is essential for template-collapse-resistant 3D CT report generation. Code will be released upon acceptance.

Bailiang Jian, Jiazhen Pan, Rohit Jena et al.

Medical image registration drives quantitative analysis across organs, modalities, and patient populations. Recent deep learning methods often combine low-level "trend-driven" computational blocks from computer vision, such as large-kernel CNNs, Transformers, and state-space models, with high-level registration-specific designs like motion pyramids, correlation layers, and iterative refinement. Yet, their relative contributions remain unclear and entangled. This raises a central question: should future advances in registration focus on importing generic architectural trends or on refining domain-specific design principles? Through a modular framework spanning brain, lung, cardiac, and abdominal registration, we systematically disentangle the influence of these two paradigms. Our evaluation reveals that low-level "trend-driven" computational blocks offer only marginal or inconsistent gains, while high-level registration-specific designs consistently deliver more accurate, smoother, and more robust deformations. These domain priors significantly elevate the performance of a standard U-Net baseline, far more than variants incorporating "trend-driven" blocks, achieving an average relative improvement of $\sim3\%$. All models and experiments are released within a transparent, modular benchmark that enables plug-and-play comparison for new architectures and registration tasks (https://github.com/BailiangJ/rethink-reg). This dynamic and extensible platform establishes a common ground for reproducible and fair evaluation, inviting the community to isolate genuine methodological contributions from domain priors. Our findings advocate a shift in research emphasis: from following architectural trends to embracing domain-specific design principles as the true drivers of progress in learning-based medical image registration.

Christian Wachinger, Bernhard Renger, Christopher Späth et al.

Interpreting quantitative CT biomarkers, such as organ volume and tissue attenuation, requires large-scale healthy reference distributions. However, creating these is challenging because clinical datasets are often heavily enriched with pathology. Here, we develop an evidence-grounded, cross-verified large language model (LLM) ensemble to filter pathological findings from radiology reports, enabling the construction of pathology-reduced cohorts from over 350,000 CT examinations. Five LLMs, first, flag structure-level abnormality candidates grounded in verbatim report evidence and, second, resolve disagreements via cross-verification. Using distribution-aware generalized additive models for location, scale, and shape, we establish comprehensive whole-body reference charts for 106 anatomical structures (volumes and attenuation) across adulthood, accounting for age, sex, contrast enhancement, and acquisition parameters. Longitudinal analyses reveal structure- and contrast-dependent changes distinct from cross-sectional trends. These resources facilitate covariate-adjusted centile scoring from routine CT, supporting standardized quantitative phenotyping, multi-site imaging studies, and scalable opportunistic screening research.

Fabian Bongratz, Anne-Marie Rickmann, Sebastian Pölsterl et al.

The reconstruction of cortical surfaces from brain magnetic resonance imaging (MRI) scans is essential for quantitative analyses of cortical thickness and sulcal morphology. Although traditional and deep learning-based algorithmic pipelines exist for this purpose, they have two major drawbacks: lengthy runtimes of multiple hours (traditional) or intricate post-processing, such as mesh extraction and topology correction (deep learning-based). In this work, we address both of these issues and propose Vox2Cortex, a deep learning-based algorithm that directly yields topologically correct, three-dimensional meshes of the boundaries of the cortex. Vox2Cortex leverages convolutional and graph convolutional neural networks to deform an initial template to the densely folded geometry of the cortex represented by an input MRI scan. We show in extensive experiments on three brain MRI datasets that our meshes are as accurate as the ones reconstructed by state-of-the-art methods in the field, without the need for time- and resource-intensive post-processing. To accurately reconstruct the tightly folded cortex, we work with meshes containing about 168,000 vertices at test time, scaling deep explicit reconstruction methods to a new level.

Morteza Ghahremani, Christian Wachinger

Recent years have witnessed a surge of interest in integrating high-dimensional data captured by multisource sensors, driven by the impressive success of neural networks in the integration of multimodal data. However, the integration of heterogeneous multimodal data poses a significant challenge, as confounding effects and dependencies among such heterogeneous data sources introduce unwanted variability and bias, leading to suboptimal performance of multimodal models. Therefore, it becomes crucial to normalize the low- or high-level features extracted from data modalities before their fusion takes place. This paper introduces a novel approach for the normalization of multimodal data, called RegBN, that incorporates regularization. RegBN uses the Frobenius norm as a regularizer term to address the side effects of confounders and underlying dependencies among different data sources. The proposed method generalizes well across multiple modalities and eliminates the need for learnable parameters, simplifying training and inference. We validate the effectiveness of RegBN on eight databases from five research areas, encompassing diverse modalities such as language, audio, image, video, depth, tabular, and 3D MRI. The proposed method demonstrates broad applicability across different architectures such as multilayer perceptrons, convolutional neural networks, and vision transformers, enabling effective normalization of both low- and high-level features in multimodal neural networks. RegBN is available at \url{https://github.com/mogvision/regbn}.

Marla Narazani, Ignacio Sarasua, Sebastian Pölsterl et al.

Alzheimer's Disease (AD) is the most common form of dementia and often difficult to diagnose due to the multifactorial etiology of dementia. Recent works on neuroimaging-based computer-aided diagnosis with deep neural networks (DNNs) showed that fusing structural magnetic resonance images (sMRI) and fluorodeoxyglucose positron emission tomography (FDG-PET) leads to improved accuracy in a study population of healthy controls and subjects with AD. However, this result conflicts with the established clinical knowledge that FDG-PET better captures AD-specific pathologies than sMRI. Therefore, we propose a framework for the systematic evaluation of multi-modal DNNs and critically re-evaluate single- and multi-modal DNNs based on FDG-PET and sMRI for binary healthy vs. AD, and three-way healthy/mild cognitive impairment/AD classification. Our experiments demonstrate that a single-modality network using FDG-PET performs better than MRI (accuracy 0.91 vs 0.87) and does not show improvement when combined. This conforms with the established clinical knowledge on AD biomarkers, but raises questions about the true benefit of multi-modal DNNs. We argue that future work on multi-modal fusion should systematically assess the contribution of individual modalities following our proposed evaluation framework. Finally, we encourage the community to go beyond healthy vs. AD classification and focus on differential diagnosis of dementia, where fusing multi-modal image information conforms with a clinical need.

Martin Rath, Morteza Ghahremani, Yitong Li et al.

Computed tomography (CT) provides rich 3D anatomical details but is often constrained by high radiation exposure, substantial costs, and limited availability. While standard chest X-rays are cost-effective and widely accessible, they only provide 2D projections with limited pathological information. Reconstructing 3D CT volumes from 2D X-rays offers a transformative solution to increase diagnostic accessibility, yet existing methods predominantly rely on synthetic X-ray projections, limiting clinical generalization. In this work, we propose AXON, a multi-stage diffusion-based framework that reconstructs high-fidelity 3D CT volumes directly from real X-rays. AXON employs a coarse-to-fine strategy, with a Brownian Bridge diffusion model-based initial stage for global structural synthesis, followed by a ControlNet-based refinement stage for local intensity optimization. It also supports bi-planar X-ray input to mitigate depth ambiguities inherent in 2D-to-3D reconstruction. A super-resolution network is integrated to upscale the generated volumes to achieve diagnostic-grade resolution. Evaluations on both public and external datasets demonstrate that AXON significantly outperforms state-of-the-art baselines, achieving a 11.9% improvement in PSNR and a 11.0% increase in SSIM with robust generalizability across disparate clinical distributions. Our code is available at https://github.com/ai-med/AXON.

Anne-Marie Rickmann, Murong Xu, Tom Nuno Wolf et al.

The wide range of research in deep learning-based medical image segmentation pushed the boundaries in a multitude of applications. A clinically relevant problem that received less attention is the handling of scans with irregular anatomy, e.g., after organ resection. State-of-the-art segmentation models often lead to organ hallucinations, i.e., false-positive predictions of organs, which cannot be alleviated by oversampling or post-processing. Motivated by the increasing need to develop robust deep learning models, we propose HALOS for abdominal organ segmentation in MR images that handles cases after organ resection surgery. To this end, we combine missing organ classification and multi-organ segmentation tasks into a multi-task model, yielding a classification-assisted segmentation pipeline. The segmentation network learns to incorporate knowledge about organ existence via feature fusion modules. Extensive experiments on a small labeled test set and large-scale UK Biobank data demonstrate the effectiveness of our approach in terms of higher segmentation Dice scores and near-to-zero false positive prediction rate.

Yitong Li, Igor Yakushev, Dennis M. Hedderich et al.

Positron emission tomography (PET) is a widely recognized technique for diagnosing neurodegenerative diseases, offering critical functional insights. However, its high costs and radiation exposure hinder its widespread use. In contrast, magnetic resonance imaging (MRI) does not involve such limitations. While MRI also detects neurodegenerative changes, it is less sensitive for diagnosis compared to PET. To overcome such limitations, one approach is to generate synthetic PET from MRI. Recent advances in generative models have paved the way for cross-modality medical image translation; however, existing methods largely emphasize structural preservation while neglecting the critical need for pathology awareness. To address this gap, we propose PASTA, a novel image translation framework built on conditional diffusion models with enhanced pathology awareness. PASTA surpasses state-of-the-art methods by preserving both structural and pathological details through its highly interactive dual-arm architecture and multi-modal condition integration. Additionally, we introduce a novel cycle exchange consistency and volumetric generation strategy that significantly enhances PASTA's ability to produce high-quality 3D PET images. Our qualitative and quantitative results demonstrate the high quality and pathology awareness of the synthesized PET scans. For Alzheimer's diagnosis, the performance of these synthesized scans improves over MRI by 4%, almost reaching the performance of actual PET. Our code is available at https://github.com/ai-med/PASTA.

Ignacio Sarasua, Sebastian Pölsterl, Christian Wachinger

Modeling temporal changes in subcortical structures is crucial for a better understanding of the progression of Alzheimer's disease (AD). Given their flexibility to adapt to heterogeneous sequence lengths, mesh-based transformer architectures have been proposed in the past for predicting hippocampus deformations across time. However, one of the main limitations of transformers is the large amount of trainable parameters, which makes the application on small datasets very challenging. In addition, current methods do not include relevant non-image information that can help to identify AD-related patterns in the progression. To this end, we introduce CASHformer, a transformer-based framework to model longitudinal shape trajectories in AD. CASHformer incorporates the idea of pre-trained transformers as universal compute engines that generalize across a wide range of tasks by freezing most layers during fine-tuning. This reduces the number of parameters by over 90% with respect to the original model and therefore enables the application of large models on small datasets without overfitting. In addition, CASHformer models cognitive decline to reveal AD atrophy patterns in the temporal sequence. Our results show that CASHformer reduces the reconstruction error by 73% compared to previously proposed methods. Moreover, the accuracy of detecting patients progressing to AD increases by 3% with imputing missing longitudinal shape data.

Anne-Marie Rickmann, Fabian Bongratz, Sebastian Pölsterl et al.

The reconstruction of cerebral cortex surfaces from brain MRI scans is instrumental for the analysis of brain morphology and the detection of cortical thinning in neurodegenerative diseases like Alzheimer's disease (AD). Moreover, for a fine-grained analysis of atrophy patterns, the parcellation of the cortical surfaces into individual brain regions is required. For the former task, powerful deep learning approaches, which provide highly accurate brain surfaces of tissue boundaries from input MRI scans in seconds, have recently been proposed. However, these methods do not come with the ability to provide a parcellation of the reconstructed surfaces. Instead, separate brain-parcellation methods have been developed, which typically consider the cortical surfaces as given, often computed beforehand with FreeSurfer. In this work, we propose two options, one based on a graph classification branch and another based on a novel generic 3D reconstruction loss, to augment template-deformation algorithms such that the surface meshes directly come with an atlas-based brain parcellation. By combining both options with two of the latest cortical surface reconstruction algorithms, we attain highly accurate parcellations with a Dice score of 90.2 (graph classification branch) and 90.4 (novel reconstruction loss) together with state-of-the-art surfaces.

Fabian Bongratz, Anne-Marie Rickmann, Christian Wachinger

Abdominal organ segmentation from CT and MRI is an essential prerequisite for surgical planning and computer-aided navigation systems. It is challenging due to the high variability in the shape, size, and position of abdominal organs. Three-dimensional numeric representations of abdominal shapes with point-wise correspondence to a template are further important for quantitative and statistical analyses thereof. Recently, template-based surface extraction methods have shown promising advances for direct mesh reconstruction from volumetric scans. However, the generalization of these deep learning-based approaches to different organs and datasets, a crucial property for deployment in clinical environments, has not yet been assessed. We close this gap and employ template-based mesh reconstruction methods for joint liver, kidney, pancreas, and spleen segmentation. Our experiments on manually annotated CT and MRI data reveal limited generalization capabilities of previous methods to organs of different geometry and weak performance on small datasets. We alleviate these issues with a novel deep diffeomorphic mesh-deformation architecture and an improved training scheme. The resulting method, UNetFlow, generalizes well to all four organs and can be easily fine-tuned on new data. Moreover, we propose a simple registration-based post-processing that aligns voxel and mesh outputs to boost segmentation accuracy.

Yitong Li, Morteza Ghahremani, Youssef Wally et al.

Diagnosing dementia, particularly for Alzheimer's Disease (AD) and frontotemporal dementia (FTD), is complex due to overlapping symptoms. While magnetic resonance imaging (MRI) and positron emission tomography (PET) data are critical for the diagnosis, integrating these modalities in deep learning faces challenges, often resulting in suboptimal performance compared to using single modalities. Moreover, the potential of multi-modal approaches in differential diagnosis, which holds significant clinical importance, remains largely unexplored. We propose a novel framework, DiaMond, to address these issues with vision Transformers to effectively integrate MRI and PET. DiaMond is equipped with self-attention and a novel bi-attention mechanism that synergistically combine MRI and PET, alongside a multi-modal normalization to reduce redundant dependency, thereby boosting the performance. DiaMond significantly outperforms existing multi-modal methods across various datasets, achieving a balanced accuracy of 92.4% in AD diagnosis, 65.2% for AD-MCI-CN classification, and 76.5% in differential diagnosis of AD and FTD. We also validated the robustness of DiaMond in a comprehensive ablation study. The code is available at https://github.com/ai-med/DiaMond.

Jiazhen Pan, Bailiang Jian, Paul Hager et al.

Ensuring the safety and reliability of large language models (LLMs) in clinical practice is critical to prevent patient harm and promote trustworthy healthcare applications of AI. However, LLMs are advancing so rapidly that static safety benchmarks often become obsolete upon publication, yielding only an incomplete and sometimes misleading picture of model trustworthiness. We demonstrate that a Dynamic, Automatic, and Systematic (DAS) red-teaming framework that continuously stress-tests LLMs can reveal significant weaknesses of current LLMs across four safety-critical domains: robustness, privacy, bias/fairness, and hallucination. A suite of adversarial agents is applied to autonomously mutate test cases, identify/evolve unsafe-triggering strategies, and evaluate responses, uncovering vulnerabilities in real time without human intervention. Applying DAS to 15 proprietary and open-source LLMs revealed a stark contrast between static benchmark performance and vulnerability under adversarial pressure. Despite a median MedQA accuracy exceeding 80\%, 94\% of previously correct answers failed our dynamic robustness tests. We observed similarly high failure rates across other domains: privacy leaks were elicited in 86\% of scenarios, cognitive-bias priming altered clinical recommendations in 81\% of fairness tests, and we identified hallucination rates exceeding 66\% in widely used models. Such profound residual risks are incompatible with routine clinical practice. By converting red-teaming from a static checklist into a dynamic stress-test audit, DAS red-teaming offers the surveillance that hospitals/regulators/technology vendors require as LLMs become embedded in patient chatbots, decision-support dashboards, and broader healthcare workflows. Our framework delivers an evolvable, scalable, and reliable safeguard for the next generation of medical AI.

Yitong Li, Tom Nuno Wolf, Sebastian Pölsterl et al.

Differential diagnosis of dementia is challenging due to overlapping symptoms, with structural magnetic resonance imaging (MRI) being the primary method for diagnosis. Despite the clinical value of computer-aided differential diagnosis, research has been limited, mainly due to the absence of public datasets that contain diverse types of dementia. This leaves researchers with small in-house datasets that are insufficient for training deep neural networks (DNNs). Self-supervised learning shows promise for utilizing unlabeled MRI scans in training, but small batch sizes for volumetric brain scans make its application challenging. To address these issues, we propose Triplet Training for differential diagnosis with limited target data. It consists of three key stages: (i) self-supervised pre-training on unlabeled data with Barlow Twins, (ii) self-distillation on task-related data, and (iii) fine-tuning on the target dataset. Our approach significantly outperforms traditional training strategies, achieving a balanced accuracy of 75.6%. We further provide insights into the training process by visualizing changes in the latent space after each step. Finally, we validate the robustness of Triplet Training in terms of its individual components in a comprehensive ablation study. Our code is available at https://github.com/ai-med/TripletTraining.

Fabian Bongratz, Yitong Li, Sama Elbaroudy et al.

Synthetic neuroimaging data can mitigate critical limitations of real-world datasets, including the scarcity of rare phenotypes, domain shifts across scanners, and insufficient longitudinal coverage. However, existing generative models largely rely on weak conditioning signals, such as labels or text, which lack anatomical grounding and often produce biologically implausible outputs. To this end, we introduce Cor2Vox, a cortex-grounded generative framework for brain magnetic resonance image (MRI) synthesis that ties image generation to continuous structural priors of the cerebral cortex. It leverages high-resolution cortical surfaces to guide a 3D shape-to-image Brownian bridge diffusion process, enabling topologically faithful synthesis and precise control over underlying anatomies. To support the generation of new, realistic brain shapes, we developed a large-scale statistical shape model of cortical morphology derived from over 33,000 UK Biobank scans. We validated the fidelity of Cor2Vox based on traditional image quality metrics, advanced cortical surface reconstruction, and whole-brain segmentation quality, outperforming many baseline methods. Across three applications, namely (i) anatomically consistent synthesis, (ii) simulation of progressive gray matter atrophy, and (iii) harmonization of in-house frontotemporal dementia scans with public datasets, Cor2Vox preserved fine-grained cortical morphology at the sub-voxel level, exhibiting remarkable robustness to variations in cortical geometry and disease phenotype without retraining.

Fabian Bongratz, Yitong Li, Sama Elbaroudy et al.

Despite recent advances in medical image generation, existing methods struggle to produce anatomically plausible 3D structures. In synthetic brain magnetic resonance images (MRIs), characteristic fissures are often missing, and reconstructed cortical surfaces appear scattered rather than densely convoluted. To address this issue, we introduce Cor2Vox, the first diffusion model-based method that translates continuous cortical shape priors to synthetic brain MRIs. To achieve this, we leverage a Brownian bridge process which allows for direct structured mapping between shape contours and medical images. Specifically, we adapt the concept of the Brownian bridge diffusion model to 3D and extend it to embrace various complementary shape representations. Our experiments demonstrate significant improvements in the geometric accuracy of reconstructed structures compared to previous voxel-based approaches. Moreover, Cor2Vox excels in image quality and diversity, yielding high variation in non-target structures like the skull. Finally, we highlight the capability of our approach to simulate cortical atrophy at the sub-voxel level. Our code is available at https://github.com/ai-med/Cor2Vox.

Yitong Li, Ralph Buchert, Benita Schmitz-Koep et al.

Positron emission tomography (PET) with 18F-Fluorodeoxyglucose (FDG) is an established tool in the diagnostic workup of patients with suspected dementing disorders. However, compared to the routinely available magnetic resonance imaging (MRI), FDG-PET remains significantly less accessible and substantially more expensive. Here, we present SiM2P, a 3D diffusion bridge-based framework that learns a probabilistic mapping from MRI and auxiliary patient information to simulate FDG-PET images of diagnostic quality. In a blinded clinical reader study, two neuroradiologists and two nuclear medicine physicians rated the original MRI and SiM2P-simulated PET images of patients with Alzheimer's disease, behavioral-variant frontotemporal dementia, and cognitively healthy controls. SiM2P significantly improved the overall diagnostic accuracy of differentiating between three groups from 75.0% to 84.7% (p<0.05). Notably, the simulated PET images received higher diagnostic certainty ratings and achieved superior interrater agreement compared to the MRI images. Finally, we developed a practical workflow for local deployment of the SiM2P framework. It requires as few as 20 site-specific cases and only basic demographic information. This approach makes the established diagnostic benefits of FDG-PET imaging more accessible to patients with suspected dementing disorders, potentially improving early detection and differential diagnosis in resource-limited settings. Our code is available at https://github.com/Yiiitong/SiM2P.

Yitong Li, Morteza Ghahremani, Christian Wachinger

Recent vision-language foundation models deliver state-of-the-art results on natural image classification but falter on medical images due to pronounced domain shifts. At the same time, training a medical foundation model requires substantial resources, including extensive annotated data and high computational capacity. To bridge this gap with minimal overhead, we introduce MedBridge, a lightweight multimodal adaptation framework that re-purposes pretrained VLMs for accurate medical image diagnosis. MedBridge comprises three key components. First, a Focal Sampling module that extracts high-resolution local regions to capture subtle pathological features and compensate for the limited input resolution of general-purpose VLMs. Second, a Query Encoder (QEncoder) injects a small set of learnable queries that attend to the frozen feature maps of VLM, aligning them with medical semantics without retraining the entire backbone. Third, a Mixture of Experts mechanism, driven by learnable queries, harnesses the complementary strength of diverse VLMs to maximize diagnostic performance. We evaluate MedBridge on five medical imaging benchmarks across three key adaptation tasks, demonstrating its superior performance in both cross-domain and in-domain adaptation settings, even under varying levels of training data availability. Notably, MedBridge achieved over 6-15% improvement in AUC compared to state-of-the-art VLM adaptation methods in multi-label thoracic disease diagnosis, underscoring its effectiveness in leveraging foundation models for accurate and data-efficient medical diagnosis. Our code is available at https://github.com/ai-med/MedBridge.

Jiajun Wang, Morteza Ghahremani, Yitong Li et al.

Controllable text-to-image (T2I) diffusion models have shown impressive performance in generating high-quality visual content through the incorporation of various conditions. Current methods, however, exhibit limited performance when guided by skeleton human poses, especially in complex pose conditions such as side or rear perspectives of human figures. To address this issue, we present Stable-Pose, a novel adapter model that introduces a coarse-to-fine attention masking strategy into a vision Transformer (ViT) to gain accurate pose guidance for T2I models. Stable-Pose is designed to adeptly handle pose conditions within pre-trained Stable Diffusion, providing a refined and efficient way of aligning pose representation during image synthesis. We leverage the query-key self-attention mechanism of ViTs to explore the interconnections among different anatomical parts in human pose skeletons. Masked pose images are used to smoothly refine the attention maps based on target pose-related features in a hierarchical manner, transitioning from coarse to fine levels. Additionally, our loss function is formulated to allocate increased emphasis to the pose region, thereby augmenting the model's precision in capturing intricate pose details. We assessed the performance of Stable-Pose across five public datasets under a wide range of indoor and outdoor human pose scenarios. Stable-Pose achieved an AP score of 57.1 in the LAION-Human dataset, marking around 13% improvement over the established technique ControlNet. The project link and code is available at https://github.com/ai-med/StablePose.

Sebastian Pölsterl, Tom Nuno Wolf, Christian Wachinger

Prior work on diagnosing Alzheimer's disease from magnetic resonance images of the brain established that convolutional neural networks (CNNs) can leverage the high-dimensional image information for classifying patients. However, little research focused on how these models can utilize the usually low-dimensional tabular information, such as patient demographics or laboratory measurements. We introduce the Dynamic Affine Feature Map Transform (DAFT), a general-purpose module for CNNs that dynamically rescales and shifts the feature maps of a convolutional layer, conditional on a patient's tabular clinical information. We show that DAFT is highly effective in combining 3D image and tabular information for diagnosis and time-to-dementia prediction, where it outperforms competing CNNs with a mean balanced accuracy of 0.622 and mean c-index of 0.748, respectively. Our extensive ablation study provides valuable insights into the architectural properties of DAFT. Our implementation is available at https://github.com/ai-med/DAFT.

Anne-Marie Rickmann, Abhijit Guha Roy, Ignacio Sarasua et al.

Fully Convolutional Neural Networks (F-CNNs) achieve state-of-the-art performance for segmentation tasks in computer vision and medical imaging. Recently, computational blocks termed squeeze and excitation (SE) have been introduced to recalibrate F-CNN feature maps both channel- and spatial-wise, boosting segmentation performance while only minimally increasing the model complexity. So far, the development of SE blocks has focused on 2D architectures. For volumetric medical images, however, 3D F-CNNs are a natural choice. In this article, we extend existing 2D recalibration methods to 3D and propose a generic compress-process-recalibrate pipeline for easy comparison of such blocks. We further introduce Project & Excite (PE) modules, customized for 3D networks. In contrast to existing modules, Project \& Excite does not perform global average pooling but compresses feature maps along different spatial dimensions of the tensor separately to retain more spatial information that is subsequently used in the excitation step. We evaluate the modules on two challenging tasks, whole-brain segmentation of MRI scans and whole-body segmentation of CT scans. We demonstrate that PE modules can be easily integrated into 3D F-CNNs, boosting performance up to 0.3 in Dice Score and outperforming 3D extensions of other recalibration blocks, while only marginally increasing the model complexity. Our code is publicly available on https://github.com/ai-med/squeeze_and_excitation .

Christian Wachinger, Anna Rieckmann, Sebastian Pölsterl

The desire to train complex machine learning algorithms and to increase the statistical power in association studies drives neuroimaging research to use ever-larger datasets. The most obvious way to increase sample size is by pooling scans from independent studies. However, simple pooling is often ill-advised as selection, measurement, and confounding biases may creep in and yield spurious correlations. In this work, we combine 35,320 magnetic resonance images of the brain from 17 studies to examine bias in neuroimaging. In the first experiment, Name That Dataset, we provide empirical evidence for the presence of bias by showing that scans can be correctly assigned to their respective dataset with 71.5% accuracy. Given such evidence, we take a closer look at confounding bias, which is often viewed as the main shortcoming in observational studies. In practice, we neither know all potential confounders nor do we have data on them. Hence, we model confounders as unknown, latent variables. Kolmogorov complexity is then used to decide whether the confounded or the causal model provides the simplest factorization of the graphical model. Finally, we present methods for dataset harmonization and study their ability to remove bias in imaging features. In particular, we propose an extension of the recently introduced ComBat algorithm to control for global variation across image features, inspired by adjusting for population stratification in genetics. Our results demonstrate that harmonization can reduce dataset-specific information in image features. Further, confounding bias can be reduced and even turned into a causal relationship. However, harmonziation also requires caution as it can easily remove relevant subject-specific information. Code is available at https://github.com/ai-med/Dataset-Bias.

Anne-Marie Rickmann, Abhijit Guha Roy, Ignacio Sarasua et al.

Fully Convolutional Neural Networks (F-CNNs) achieve state-of-the-art performance for image segmentation in medical imaging. Recently, squeeze and excitation (SE) modules and variations thereof have been introduced to recalibrate feature maps channel- and spatial-wise, which can boost performance while only minimally increasing model complexity. So far, the development of SE has focused on 2D images. In this paper, we propose `Project & Excite' (PE) modules that base upon the ideas of SE and extend them to operating on 3D volumetric images. `Project & Excite' does not perform global average pooling, but squeezes feature maps along different slices of a tensor separately to retain more spatial information that is subsequently used in the excitation step. We demonstrate that PE modules can be easily integrated in 3D U-Net, boosting performance by 5% Dice points, while only increasing the model complexity by 2%. We evaluate the PE module on two challenging tasks, whole-brain segmentation of MRI scans and whole-body segmentation of CT scans. Code: https://github.com/ai-med/squeeze_and_excitation

Sebastian Pölsterl, Christian Wachinger

Recent methods for generating novel molecules use graph representations of molecules and employ various forms of graph convolutional neural networks for inference. However, training requires solving an expensive graph isomorphism problem, which previous approaches do not address or solve only approximately. In this work, we propose ALMGIG, a likelihood-free adversarial learning framework for inference and de novo molecule generation that avoids explicitly computing a reconstruction loss. Our approach extends generative adversarial networks by including an adversarial cycle-consistency loss to implicitly enforce the reconstruction property. To capture properties unique to molecules, such as valence, we extend the Graph Isomorphism Network to multi-graphs. To quantify the performance of models, we propose to compute the distance between distributions of physicochemical properties with the 1-Wasserstein distance. We demonstrate that ALMGIG more accurately learns the distribution over the space of molecules than all baselines. Moreover, it can be utilized for drug discovery by efficiently searching the space of molecules using molecules' continuous latent representation. Our code is available at https://github.com/ai-med/almgig

Abhijit Guha Roy, Shayan Siddiqui, Sebastian Pölsterl et al.

Deep neural networks enable highly accurate image segmentation, but require large amounts of manually annotated data for supervised training. Few-shot learning aims to address this shortcoming by learning a new class from a few annotated support examples. We introduce, a novel few-shot framework, for the segmentation of volumetric medical images with only a few annotated slices. Compared to other related works in computer vision, the major challenges are the absence of pre-trained networks and the volumetric nature of medical scans. We address these challenges by proposing a new architecture for few-shot segmentation that incorporates 'squeeze & excite' blocks. Our two-armed architecture consists of a conditioner arm, which processes the annotated support input and generates a task-specific representation. This representation is passed on to the segmenter arm that uses this information to segment the new query image. To facilitate efficient interaction between the conditioner and the segmenter arm, we propose to use 'channel squeeze & spatial excitation' blocks - a light-weight computational module - that enables heavy interaction between both the arms with negligible increase in model complexity. This contribution allows us to perform image segmentation without relying on a pre-trained model, which generally is unavailable for medical scans. Furthermore, we propose an efficient strategy for volumetric segmentation by optimally pairing a few slices of the support volume to all the slices of the query volume. We perform experiments for organ segmentation on whole-body contrast-enhanced CT scans from the Visceral Dataset. Our proposed model outperforms multiple baselines and existing approaches with respect to the segmentation accuracy by a significant margin. The source code is available at https://github.com/abhi4ssj/few-shot-segmentation.

Fabian Bongratz, Anne-Marie Rickmann, Christian Wachinger

The reconstruction of cortical surfaces is a prerequisite for quantitative analyses of the cerebral cortex in magnetic resonance imaging (MRI). Existing segmentation-based methods separate the surface registration from the surface extraction, which is computationally inefficient and prone to distortions. We introduce Vox2Cortex-Flow (V2C-Flow), a deep mesh-deformation technique that learns a deformation field from a brain template to the cortical surfaces of an MRI scan. To this end, we present a geometric neural network that models the deformation-describing ordinary differential equation in a continuous manner. The network architecture comprises convolutional and graph-convolutional layers, which allows it to work with images and meshes at the same time. V2C-Flow is not only very fast, requiring less than two seconds to infer all four cortical surfaces, but also establishes vertex-wise correspondences to the template during reconstruction. In addition, V2C-Flow is the first approach for cortex reconstruction that models white matter and pial surfaces jointly, therefore avoiding intersections between them. Our comprehensive experiments on internal and external test data demonstrate that V2C-Flow results in cortical surfaces that are state-of-the-art in terms of accuracy. Moreover, we show that the established correspondences are more consistent than in FreeSurfer and that they can directly be utilized for cortex parcellation and group analyses of cortical thickness.

Tom Nuno Wolf, Fabian Bongratz, Anne-Marie Rickmann et al.

Explaining predictions of black-box neural networks is crucial when applied to decision-critical tasks. Thus, attribution maps are commonly used to identify important image regions, despite prior work showing that humans prefer explanations based on similar examples. To this end, ProtoPNet learns a set of class-representative feature vectors (prototypes) for case-based reasoning. During inference, similarities of latent features to prototypes are linearly classified to form predictions and attribution maps are provided to explain the similarity. In this work, we evaluate whether architectures for case-based reasoning fulfill established axioms required for faithful explanations using the example of ProtoPNet. We show that such architectures allow the extraction of faithful explanations. However, we prove that the attribution maps used to explain the similarities violate the axioms. We propose a new procedure to extract explanations for trained ProtoPNets, named ProtoPFaith. Conceptually, these explanations are Shapley values, calculated on the similarity scores of each prototype. They allow to faithfully answer which prototypes are present in an unseen image and quantify each pixel's contribution to that presence, thereby complying with all axioms. The theoretical violations of ProtoPNet manifest in our experiments on three datasets (CUB-200-2011, Stanford Dogs, RSNA) and five architectures (ConvNet, ResNet, ResNet50, WideResNet50, ResNeXt50). Our experiments show a qualitative difference between the explanations given by ProtoPNet and ProtoPFaith. Additionally, we quantify the explanations with the Area Over the Perturbation Curve, on which ProtoPFaith outperforms ProtoPNet on all experiments by a factor $>10^3$.

Che Liu, Yinda Chen, Haoyuan Shi et al.

The advent of large-scale vision foundation models, pre-trained on diverse natural images, has marked a paradigm shift in computer vision. However, how the frontier vision foundation models' efficacies transfer to specialized domains remains such as medical imaging remains an open question. This report investigates whether DINOv3, a state-of-the-art self-supervised vision transformer (ViT) that features strong capability in dense prediction tasks, can directly serve as a powerful, unified encoder for medical vision tasks without domain-specific pre-training. To answer this, we benchmark DINOv3 across common medical vision tasks, including 2D/3D classification and segmentation on a wide range of medical imaging modalities. We systematically analyze its scalability by varying model sizes and input image resolutions. Our findings reveal that DINOv3 shows impressive performance and establishes a formidable new baseline. Remarkably, it can even outperform medical-specific foundation models like BiomedCLIP and CT-Net on several tasks, despite being trained solely on natural images. However, we identify clear limitations: The model's features degrade in scenarios requiring deep domain specialization, such as in Whole-Slide Pathological Images (WSIs), Electron Microscopy (EM), and Positron Emission Tomography (PET). Furthermore, we observe that DINOv3 does not consistently obey scaling law in the medical domain; performance does not reliably increase with larger models or finer feature resolutions, showing diverse scaling behaviors across tasks. Ultimately, our work establishes DINOv3 as a strong baseline, whose powerful visual features can serve as a robust prior for multiple complex medical tasks. This opens promising future directions, such as leveraging its features to enforce multiview consistency in 3D reconstruction.

Bailiang Jian, Jiazhen Pan, Yitong Li et al.

Longitudinal brain analysis is essential for understanding healthy aging and identifying pathological deviations. Longitudinal registration of sequential brain MRI underpins such analyses. However, existing methods are limited by reliance on densely sampled time series, a trade-off between accuracy and temporal smoothness, and an inability to prospectively forecast future brain states. To overcome these challenges, we introduce \emph{TimeFlow}, a learning-based framework for longitudinal brain MRI registration. TimeFlow uses a U-Net backbone with temporal conditioning to model neuroanatomy as a continuous function of age. Given only two scans from an individual, TimeFlow estimates accurate and temporally coherent deformation fields, enabling non-linear extrapolation to predict future brain states. This is achieved by our proposed inter-/extra-polation consistency constraints applied to both the deformation fields and deformed images. Remarkably, these constraints preserve temporal consistency and continuity without requiring explicit smoothness regularizers or densely sampled sequential data. Extensive experiments demonstrate that TimeFlow outperforms state-of-the-art methods in terms of both future timepoint forecasting and registration accuracy. Moreover, TimeFlow supports novel biological brain aging analyses by differentiating neurodegenerative trajectories from normal aging without requiring segmentation, thereby eliminating the need for labor-intensive annotations and mitigating segmentation inconsistency. TimeFlow offers an accurate, data-efficient, and annotation-free framework for longitudinal analysis of brain aging and chronic diseases, capable of forecasting brain changes beyond the observed study period.

Fabian Bongratz, Jan Fecht, Anne-Marie Rickmann et al.

Reconstructing the cortex from longitudinal magnetic resonance imaging (MRI) is indispensable for analyzing morphological alterations in the human brain. Despite the recent advancement of cortical surface reconstruction with deep learning, challenges arising from longitudinal data are still persistent. Especially the lack of strong spatiotemporal point correspondence between highly convoluted brain surfaces hinders downstream analyses, as local morphology is not directly comparable if the anatomical location is not matched precisely. To address this issue, we present V2C-Long, the first dedicated deep learning-based cortex reconstruction method for longitudinal MRI. V2C-Long exhibits strong inherent spatiotemporal correspondence across subjects and visits, thereby reducing the need for surface-based post-processing. We establish this correspondence directly during the reconstruction via the composition of two deep template-deformation networks and innovative aggregation of within-subject templates in mesh space. We validate V2C-Long on two large neuroimaging studies, focusing on surface accuracy, consistency, generalization, test-retest reliability, and sensitivity. The results reveal a substantial improvement in longitudinal consistency and accuracy compared to existing methods. In addition, we demonstrate stronger evidence for longitudinal cortical atrophy in Alzheimer's disease than longitudinal FreeSurfer.

Patrick Madlindl, Fabian Bongratz, Christian Wachinger

Cortical surface reconstruction (CSR) from magnetic resonance imaging (MRI) is fundamental to neuroimage analysis, enabling morphological studies of the cerebral cortex and functional brain mapping. Recent advances in learning-based CSR have dramatically accelerated processing, allowing for reconstructions through the deformation of anatomical templates within seconds. However, ensuring the learned deformations are optimal in terms of deformation energy and consistent across training runs remains a particular challenge. In this work, we design a Minimal Energy Deformation (MED) loss, acting as a regularizer on the deformation trajectories and complementing the widely used Chamfer distance in CSR. We incorporate it into the recent V2C-Flow model and demonstrate considerable improvements in previously neglected training consistency and reproducibility without harming reconstruction accuracy and topological correctness.

Ivan Stoyanov, Fabian Bongratz, Christian Wachinger

Accurate forecasting of individualized, high-resolution cortical thickness (CTh) trajectories is essential for detecting subtle cortical changes, providing invaluable insights into neurodegenerative processes and facilitating earlier and more precise intervention strategies. However, CTh forecasting is a challenging task due to the intricate non-Euclidean geometry of the cerebral cortex and the need to integrate multi-modal data for subject-specific predictions. To address these challenges, we introduce the Spherical Brownian Bridge Diffusion Model (SBDM). Specifically, we propose a bidirectional conditional Brownian bridge diffusion process to forecast CTh trajectories at the vertex level of registered cortical surfaces. Our technical contribution includes a new denoising model, the conditional spherical U-Net (CoS-UNet), which combines spherical convolutions and dense cross-attention to integrate cortical surfaces and tabular conditions seamlessly. Compared to previous approaches, SBDM achieves significantly reduced prediction errors, as demonstrated by our experiments based on longitudinal datasets from the ADNI and OASIS. Additionally, we demonstrate SBDM's ability to generate individual factual and counterfactual CTh trajectories, offering a novel framework for exploring hypothetical scenarios of cortical development.

Lasse Hansen, Wiebke Heyer, Christoph Großbröhmer et al.

Medical image registration is critical for clinical applications, and fair benchmarking of different methods is essential for monitoring ongoing progress. To date, the Learn2Reg 2020-2023 challenges have released several complementary datasets and established metrics for evaluations. However, these editions did not capture all aspects of the registration problem, particularly in terms of modality diversity and task complexity. To address these limitations, the 2024 edition introduces three new tasks, including large-scale multi-modal registration and unsupervised inter-subject brain registration, as well as the first microscopy-focused benchmark within Learn2Reg. The new datasets also inspired new method developments, including invertibility constraints, pyramid features, keypoints alignment and instance optimisation.

Fabian Bongratz, Tom Nuno Wolf, Jaume Gual Ramon et al.

Interpretable models are crucial for supporting clinical decision-making, driving advances in their development and application for medical images. However, the nature of 3D volumetric data makes it inherently challenging to visualize and interpret intricate and complex structures like the cerebral cortex. Cortical surface renderings, on the other hand, provide a more accessible and understandable 3D representation of brain anatomy, facilitating visualization and interactive exploration. Motivated by this advantage and the widespread use of surface data for studying neurological disorders, we present the eXplainable Surface Vision Transformer (X-SiT). This is the first inherently interpretable neural network that offers human-understandable predictions based on interpretable cortical features. As part of X-SiT, we introduce a prototypical surface patch decoder for classifying surface patch embeddings, incorporating case-based reasoning with spatially corresponding cortical prototypes. The results demonstrate state-of-the-art performance in detecting Alzheimer's disease and frontotemporal dementia while additionally providing informative prototypes that align with known disease patterns and reveal classification errors.

Emre Kavak, Tom Nuno Wolf, Christian Wachinger

Dataset bias often leads deep learning models to exploit spurious correlations instead of task-relevant signals. We introduce the Standard Anti-Causal Model (SAM), a unifying causal framework that characterizes bias mechanisms and yields a conditional independence criterion for causal stability. Building on this theory, we propose DISCO$_m$ and sDISCO, efficient and scalable estimators of conditional distance correlation that enable independence regularization in black-box models. Across five diverse datasets, our methods consistently outperform or are competitive in existing bias mitigation approaches, while requiring fewer hyperparameters and scaling seamlessly to multi-bias scenarios. This work bridges causal theory and practical deep learning, providing both a principled foundation and effective tools for robust prediction. Source Code: https://github.com/***.

Tom Nuno Wolf, Emre Kavak, Fabian Bongratz et al.

The deployment of deep learning models in critical domains necessitates a balance between high accuracy and interpretability. We introduce SIC, an inherently interpretable neural network that provides local and global explanations of its decision-making process. Leveraging the concept of case-based reasoning, SIC extracts class-representative support vectors from training images, ensuring they capture relevant features while suppressing irrelevant ones. Classification decisions are made by calculating and aggregating similarity scores between these support vectors and the input's latent feature vector. We employ B-Cos transformations, which align model weights with inputs, to yield coherent pixel-level explanations in addition to global explanations of case-based reasoning. We evaluate SIC on three tasks: fine-grained classification on Stanford Dogs and FunnyBirds, multi-label classification on Pascal VOC, and pathology detection on the RSNA dataset. Results indicate that SIC not only achieves competitive accuracy compared to state-of-the-art black-box and inherently interpretable models but also offers insightful explanations verified through practical evaluation on the FunnyBirds benchmark. Our theoretical analysis proves that these explanations fulfill established axioms for explanations. Our findings underscore SIC's potential for applications where understanding model decisions is as critical as the decisions themselves.

Fabian Bongratz, Markus Karmann, Adrian Holz et al.

Meningeal lymphatic vessels (MLVs) are responsible for the drainage of waste products from the human brain. An impairment in their functionality has been associated with aging as well as brain disorders like multiple sclerosis and Alzheimer's disease. However, MLVs have only recently been described for the first time in magnetic resonance imaging (MRI), and their ramified structure renders manual segmentation particularly difficult. Further, as there is no consistent notion of their appearance, human-annotated MLV structures contain a high inter-rater variability that most automatic segmentation methods cannot take into account. In this work, we propose a new rater-aware training scheme for the popular nnU-Net model, and we explore rater-based ensembling strategies for accurate and consistent segmentation of MLVs. This enables us to boost nnU-Net's performance while obtaining explicit predictions in different annotation styles and a rater-based uncertainty estimation. Our final model, MLV$^2$-Net, achieves a Dice similarity coefficient of 0.806 with respect to the human reference standard. The model further matches the human inter-rater reliability and replicates age-related associations with MLV volume.

Christian Wachinger, Dennis Hedderich, Melissa Thalhammer et al.

Understanding individual differences in cortical structure is key to advancing diagnostics in neurology and psychiatry. Reference models aid in detecting aberrant cortical thickness, yet site-specific biases limit their direct application to unseen data, and region-wise averages prevent the detection of localized cortical changes. To address these limitations, we developed the Stochastic Cortical Self-Reconstruction (SCSR), a novel method that leverages deep learning to reconstruct cortical thickness maps at the vertex level without needing additional subject information. Trained on over 25,000 healthy individuals, SCSR generates highly individualized cortical reconstructions that can detect subtle thickness deviations. Our evaluations on independent test sets demonstrated that SCSR achieved significantly lower reconstruction errors and identified atrophy patterns that enabled better disease discrimination than established methods. It also hints at cortical thinning in preterm infants that went undetected by existing models, showcasing its versatility. Finally, SCSR excelled in mapping highly resolved cortical deviations of dementia patients from clinical data, highlighting its potential for supporting diagnosis in clinical practice.

Ignacio Sarasua, Sebastian Pölsterl, Christian Wachinger

The longitudinal modeling of neuroanatomical changes related to Alzheimer's disease (AD) is crucial for studying the progression of the disease. To this end, we introduce TransforMesh, a spatio-temporal network based on transformers that models longitudinal shape changes on 3D anatomical meshes. While transformer and mesh networks have recently shown impressive performances in natural language processing and computer vision, their application to medical image analysis has been very limited. To the best of our knowledge, this is the first work that combines transformer and mesh networks. Our results show that TransforMesh can model shape trajectories better than other baseline architectures that do not capture temporal dependencies. Moreover, we also explore the capabilities of TransforMesh in detecting structural anomalies of the hippocampus in patients developing AD.

Raphael Ronge, Kwangsik Nho, Christian Wachinger et al.

The current state-of-the-art deep neural networks (DNNs) for Alzheimer's Disease diagnosis use different biomarker combinations to classify patients, but do not allow extracting knowledge about the interactions of biomarkers. However, to improve our understanding of the disease, it is paramount to extract such knowledge from the learned model. In this paper, we propose a Deep Factorization Machine model that combines the ability of DNNs to learn complex relationships and the ease of interpretability of a linear model. The proposed model has three parts: (i) an embedding layer to deal with sparse categorical data, (ii) a Factorization Machine to efficiently learn pairwise interactions, and (iii) a DNN to implicitly model higher order interactions. In our experiments on data from the Alzheimer's Disease Neuroimaging Initiative, we demonstrate that our proposed model classifies cognitive normal, mild cognitive impaired, and demented patients more accurately than competing models. In addition, we show that valuable knowledge about the interactions among biomarkers can be obtained.

Sebastian Pölsterl, Christina Aigner, Christian Wachinger

Deep Neural Networks (DNNs) have an enormous potential to learn from complex biomedical data. In particular, DNNs have been used to seamlessly fuse heterogeneous information from neuroanatomy, genetics, biomarkers, and neuropsychological tests for highly accurate Alzheimer's disease diagnosis. On the other hand, their black-box nature is still a barrier for the adoption of such a system in the clinic, where interpretability is absolutely essential. We propose Shapley Value Explanation of Heterogeneous Neural Networks (SVEHNN) for explaining the Alzheimer's diagnosis made by a DNN from the 3D point cloud of the neuroanatomy and tabular biomarkers. Our explanations are based on the Shapley value, which is the unique method that satisfies all fundamental axioms for local explanations previously established in the literature. Thus, SVEHNN has many desirable characteristics that previous work on interpretability for medical decision making is lacking. To avoid the exponential time complexity of the Shapley value, we propose to transform a given DNN into a Lightweight Probabilistic Deep Network without re-training, thus achieving a complexity only quadratic in the number of features. In our experiments on synthetic and real data, we show that we can closely approximate the exact Shapley value with a dramatically reduced runtime and can reveal the hidden knowledge the network has learned from the data.

Michela Antonelli, Annika Reinke, Spyridon Bakas et al.

International challenges have become the de facto standard for comparative assessment of image analysis algorithms given a specific task. Segmentation is so far the most widely investigated medical image processing task, but the various segmentation challenges have typically been organized in isolation, such that algorithm development was driven by the need to tackle a single specific clinical problem. We hypothesized that a method capable of performing well on multiple tasks will generalize well to a previously unseen task and potentially outperform a custom-designed solution. To investigate the hypothesis, we organized the Medical Segmentation Decathlon (MSD) - a biomedical image analysis challenge, in which algorithms compete in a multitude of both tasks and modalities. The underlying data set was designed to explore the axis of difficulties typically encountered when dealing with medical images, such as small data sets, unbalanced labels, multi-site data and small objects. The MSD challenge confirmed that algorithms with a consistent good performance on a set of tasks preserved their good average performance on a different set of previously unseen tasks. Moreover, by monitoring the MSD winner for two years, we found that this algorithm continued generalizing well to a wide range of other clinical problems, further confirming our hypothesis. Three main conclusions can be drawn from this study: (1) state-of-the-art image segmentation algorithms are mature, accurate, and generalize well when retrained on unseen tasks; (2) consistent algorithmic performance across multiple tasks is a strong surrogate of algorithmic generalizability; (3) the training of accurate AI segmentation models is now commoditized to non AI experts.

Fabian Gröger, Anne-Marie Rickmann, Christian Wachinger

We propose an unsupervised domain adaptation (UDA) approach for white matter hyperintensity (WMH) segmentation, which uses Self-Training with Uncertainty DEpendent Label refinement (STRUDEL). Self-training has recently been introduced as a highly effective method for UDA, which is based on self-generated pseudo labels. However, pseudo labels can be very noisy and therefore deteriorate model performance. We propose to predict the uncertainty of pseudo labels and integrate it in the training process with an uncertainty-guided loss function to highlight labels with high certainty. STRUDEL is further improved by incorporating the segmentation output of an existing method in the pseudo label generation that showed high robustness for WMH segmentation. In our experiments, we evaluate STRUDEL with a standard U-Net and a modified network with a higher receptive field. Our results on WMH segmentation across datasets demonstrate the significant improvement of STRUDEL with respect to standard self-training.

Ignacio Sarasua, Jonwong Lee, Christian Wachinger

Geometric deep learning can find representations that are optimal for a given task and therefore improve the performance over pre-defined representations. While current work has mainly focused on point representations, meshes also contain connectivity information and are therefore a more comprehensive characterization of the underlying anatomical surface. In this work, we evaluate four recent geometric deep learning approaches that operate on mesh representations. These approaches can be grouped into template-free and template-based approaches, where the template-based methods need a more elaborate pre-processing step with the definition of a common reference template and correspondences. We compare the different networks for the prediction of Alzheimer's disease based on the meshes of the hippocampus. Our results show advantages for template-based methods in terms of accuracy, number of learnable parameters, and training speed. While the template creation may be limiting for some applications, neuroimaging has a long history of building templates with automated tools readily available. Overall, working with meshes is more involved than working with simplistic point clouds, but they also offer new avenues for designing geometric deep learning architectures.

Ignacio Sarasua, Sebastian Poelsterl, Christian Wachinger

Spatial and channel re-calibration have become powerful concepts in computer vision. Their ability to capture long-range dependencies is especially useful for those networks that extract local features, such as CNNs. While re-calibration has been widely studied for image analysis, it has not yet been used on shape representations. In this work, we introduce re-calibration modules on deep neural networks for 3D point clouds. We propose a set of re-calibration blocks that extend Squeeze and Excitation blocks and that can be added to any network for 3D point cloud analysis that builds a global descriptor by hierarchically combining features from multiple local neighborhoods. We run two sets of experiments to validate our approach. First, we demonstrate the benefit and versatility of our proposed modules by incorporating them into three state-of-the-art networks for 3D point cloud analysis: PointNet++, DGCNN, and RSCNN. We evaluate each network on two tasks: object classification on ModelNet40, and object part segmentation on ShapeNet. Our results show an improvement of up to 1% in accuracy for ModelNet40 compared to the baseline method. In the second set of experiments, we investigate the benefits of re-calibration blocks on Alzheimer's Disease (AD) diagnosis. Our results demonstrate that our proposed methods yield a 2% increase in accuracy for diagnosing AD and a 2.3% increase in concordance index for predicting AD onset with time-to-event analysis. Concluding, re-calibration improves the accuracy of point cloud architectures, while only minimally increasing the number of parameters.

Philipp Kopper, Sebastian Pölsterl, Christian Wachinger et al.

We propose a versatile framework for survival analysis that combines advanced concepts from statistics with deep learning. The presented framework is based on piecewise exponential models and thereby supports various survival tasks, such as competing risks and multi-state modeling, and further allows for estimation of time-varying effects and time-varying features. To also include multiple data sources and higher-order interaction effects into the model, we embed the model class in a neural network and thereby enable the simultaneous estimation of both inherently interpretable structured regression inputs as well as deep neural network components which can potentially process additional unstructured data sources. A proof of concept is provided by using the framework to predict Alzheimer's disease progression based on tabular and 3D point cloud data and applying it to synthetic data.