arXiv subjects

Nils D. Forkert

Publications and source records attributed to Nils D. Forkert.

At least 19 recordsLinked to original sources

arXiv2026

Deep Learning for Retinal Degeneration Assessment: A Comprehensive Analysis of the MARIO Challenge

The MARIO challenge, held at MICCAI 2024, focused on advancing the automated detection and monitoring of age-related macular degeneration (AMD) through the analysis of optical coherence tomography (OCT) images. Designed to evaluate algorithmic performance in detecting neovascular activity changes within AMD, the challenge incorporated unique multi-modal datasets. The primary dataset, sourced from Brest, France, was used by participating teams to train and test their models. The final ranking was determined based on performance on this dataset. An auxiliary dataset from Algeria was used post-challenge to evaluate population and device shifts from submitted solutions. Two tasks were involved in the MARIO challenge. The first one was the classification of evolution between two consecutive 2D OCT B-scans. The second one was the prediction of future AMD evolution over three months for patients undergoing anti-vascular endothelial growth factor (VEGF) therapy. Thirty-five teams participated, with the top 12 finalists presenting their methods. This paper outlines the challenge's structure, tasks, data characteristics, and winning methodologies, setting a benchmark for AMD monitoring using OCT, infrared imaging, and clinical data (such as the number of visits, age, gender, etc.). The results of this challenge indicate that artificial intelligence (AI) performs as well as a physician in measuring AMD progression (Task 1) but is not yet able of predicting future evolution (Task 2).

cs.CV↗

arXiv2026

A Neuroimaging Simulation Framework for Developing and Evaluating Causal AI

Causally linking disease-related factors to image-derived biomarkers provides a powerful pathway to understanding disease mechanisms. Despite growing interest in applying causal artificial intelligence (AI) approaches for this task, these methods still need to be adapted for complex medical images, and especially, neuroimaging. However, the lack of ground-truth data presents a barrier to development. To bridge this gap, we developed and tested a method for generating synthetic neuroimages, which adhere to a user-specified causal structure describing the non-image to image variable relationships, permitting the creation of ground-truth neuroimaging datasets. In the simulated T1-weighted magnetic resonance images, anatomical variability is modeled by sampling from a subspace estimated from real data and deforming a template image to create unique simulated subjects. Causal relationships are encoded via precise volumetric changes of any region-of-interest without unwanted global artifacts. We achieved relative volume errors of 0.3-2.66% for the targeted regions-of-interest and demonstrate their statistically significant causal relationships, while maintaining mean absolute errors for non-target brain regions between 0.034-0.397ml. An initial evaluation of causal discovery methods exposes their limited ability to suppress spurious connections, highlighting the need for image-appropriate methods. Our framework is the first to enable the generation of realistic synthetic 3D neuroimages with explicit causal control that can serve as the missing ground-truth data necessary for the objective benchmarking and development of causal AI methods.

eess.IV↗

arXiv2026

Towards Brain MRI Foundation Models for the Clinic: Findings from the FOMO25 Challenge

Clinical deployment of automated brain MRI analysis faces a fundamental challenge: clinical data is heterogeneous and noisy, and high-quality labels are prohibitively costly to obtain. Self-supervised learning (SSL) can address this by leveraging the vast amounts of unlabeled data produced in clinical workflows to train robust \textit{foundation models} that adapt out-of-domain with minimal supervision. However, the development of foundation models for brain MRI has been limited by small pretraining datasets and in-domain benchmarking focused on high-quality, research-grade data. To address this gap, we organized the FOMO25 challenge as a satellite event at MICCAI 2025. FOMO25 provided participants with a large pretraining dataset, FOMO60K, and evaluated models on data sourced directly from clinical workflows in few-shot and out-of-domain settings. Tasks covered infarct classification, meningioma segmentation, and brain age regression, and considered both models trained on FOMO60K (method track) and any data (open track). Nineteen foundation models from sixteen teams were evaluated using a standardized containerized pipeline. Results show that (a) self-supervised pretraining improves generalization on clinical data under domain shift, with the strongest models trained \textit{out-of-domain} surpassing supervised baselines trained \textit{in-domain}. (b) No single pretraining objective benefits all tasks: MAE favors segmentation, hybrid reconstruction-contrastive objectives favor classification, and (c) strong performance was achieved by small pretrained models, and improvements from scaling model size and training duration did not yield reliable benefits.

cs.CV↗

arXiv2026

CRAwDAD: Causal Reasoning Augmentation with Dual-Agent Debate

When people reason about cause and effect, they often consider many competing "what if" scenarios before deciding which explanation fits best. Analogously, advanced language models capable of causal inference can consider multiple interventions and counterfactuals to judge the validity of causal claims. Crucially, this type of reasoning is less like a single calculation and more like an internal dialogue between alternative hypotheses. In this paper, we make this dialogue explicit through a dual-agent debate framework where one model provides a structured causal inference, and the other critically examines this reasoning for logical flaws. When disagreements arise, the agents attempt to persuade each other, challenging each other's logic and revising their conclusions until they converge on a mutually agreed answer. To take advantage of this deliberative process, we specifically use reasoning language models, whose strengths in both causal inference and adversarial debate remain under-explored relative to standard large language models. We evaluate our approach on the CLadder dataset, a benchmark linking natural language questions to formally defined causal graphs across all three rungs of Pearl's ladder of causation. With Qwen3 and DeepSeek-R1 as debater agents, we demonstrate that multi-agent debate improves DeepSeek-R1's overall accuracy in causal inference from 78.03% to 87.45%, with the counterfactual category specifically improving from 67.94% to 80.04% accuracy. Similarly, Qwen3's overall accuracy improves from 84.16% to 89.41%, and counterfactual questions from 71.53% to 80.35%, showing that even strong models can still benefit greatly from debate with weaker agents. Our results highlight the potential of reasoning models as building blocks for multi-agent systems in causal inference, and demonstrate the importance of diverse perspectives in causal problem-solving.

cs.LG↗

arXiv2025

Semi-disentangled spatiotemporal implicit neural representations of longitudinal neuroimaging data for trajectory classification

The human brain undergoes dynamic, potentially pathology-driven, structural changes throughout a lifespan. Longitudinal Magnetic Resonance Imaging (MRI) and other neuroimaging data are valuable for characterizing trajectories of change associated with typical and atypical aging. However, the analysis of such data is highly challenging given their discrete nature with different spatial and temporal image sampling patterns within individuals and across populations. This leads to computational problems for most traditional deep learning methods that cannot represent the underlying continuous biological process. To address these limitations, we present a new, fully data-driven method for representing aging trajectories across the entire brain by modelling subject-specific longitudinal T1-weighted MRI data as continuous functions using Implicit Neural Representations (INRs). Therefore, we introduce a novel INR architecture capable of partially disentangling spatial and temporal trajectory parameters and design an efficient framework that directly operates on the INRs' parameter space to classify brain aging trajectories. To evaluate our method in a controlled data environment, we develop a biologically grounded trajectory simulation and generate T1-weighted 3D MRI data for 450 healthy and dementia-like subjects at regularly and irregularly sampled timepoints. In the more realistic irregular sampling experiment, our INR-based method achieves 81.3% accuracy for the brain aging trajectory classification task, outperforming a standard deep learning baseline model (73.7%).

cs.CV↗

arXiv2025

Exploring the interplay of label bias with subgroup size and separability: A case study in mammographic density classification

Systematic mislabelling affecting specific subgroups (i.e., label bias) in medical imaging datasets represents an understudied issue concerning the fairness of medical AI systems. In this work, we investigated how size and separability of subgroups affected by label bias influence the learned features and performance of a deep learning model. Therefore, we trained deep learning models for binary tissue density classification using the EMory BrEast imaging Dataset (EMBED), where label bias affected separable subgroups (based on imaging manufacturer) or non-separable "pseudo-subgroups". We found that simulated subgroup label bias led to prominent shifts in the learned feature representations of the models. Importantly, these shifts within the feature space were dependent on both the relative size and the separability of the subgroup affected by label bias. We also observed notable differences in subgroup performance depending on whether a validation set with clean labels was used to define the classification threshold for the model. For instance, with label bias affecting the majority separable subgroup, the true positive rate for that subgroup fell from 0.898, when the validation set had clean labels, to 0.518, when the validation set had biased labels. Our work represents a key contribution toward understanding the consequences of label bias on subgroup fairness in medical imaging AI.

cs.CV↗

arXiv2025

ISLES'24: Final Infarct Prediction with Multimodal Imaging and Clinical Data. Where Do We Stand?

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.

eess.IV↗

arXiv2024

Towards objective and systematic evaluation of bias in artificial intelligence for medical imaging

Artificial intelligence (AI) models trained using medical images for clinical tasks often exhibit bias in the form of disparities in performance between subgroups. Since not all sources of biases in real-world medical imaging data are easily identifiable, it is challenging to comprehensively assess how those biases are encoded in models, and how capable bias mitigation methods are at ameliorating performance disparities. In this article, we introduce a novel analysis framework for systematically and objectively investigating the impact of biases in medical images on AI models. We developed and tested this framework for conducting controlled in silico trials to assess bias in medical imaging AI using a tool for generating synthetic magnetic resonance images with known disease effects and sources of bias. The feasibility is showcased by using three counterfactual bias scenarios to measure the impact of simulated bias effects on a convolutional neural network (CNN) classifier and the efficacy of three bias mitigation strategies. The analysis revealed that the simulated biases resulted in expected subgroup performance disparities when the CNN was trained on the synthetic datasets. Moreover, reweighing was identified as the most successful bias mitigation strategy for this setup, and we demonstrated how explainable AI methods can aid in investigating the manifestation of bias in the model using this framework. Developing fair AI models is a considerable challenge given that many and often unknown sources of biases can be present in medical imaging datasets. In this work, we present a novel methodology to objectively study the impact of biases and mitigation strategies on deep learning pipelines, which can support the development of clinical AI that is robust and responsible.

cs.CV↗

arXiv2022

Weakly Supervised Medical Image Segmentation With Soft Labels and Noise Robust Loss

Recent advances in deep learning algorithms have led to significant benefits for solving many medical image analysis problems. Training deep learning models commonly requires large datasets with expert-labeled annotations. However, acquiring expert-labeled annotation is not only expensive but also is subjective, error-prone, and inter-/intra- observer variability introduces noise to labels. This is particularly a problem when using deep learning models for segmenting medical images due to the ambiguous anatomical boundaries. Image-based medical diagnosis tools using deep learning models trained with incorrect segmentation labels can lead to false diagnoses and treatment suggestions. Multi-rater annotations might be better suited to train deep learning models with small training sets compared to single-rater annotations. The aim of this paper was to develop and evaluate a method to generate probabilistic labels based on multi-rater annotations and anatomical knowledge of the lesion features in MRI and a method to train segmentation models using probabilistic labels using normalized active-passive loss as a "noise-tolerant loss" function. The model was evaluated by comparing it to binary ground truth for 17 knees MRI scans for clinical segmentation and detection of bone marrow lesions (BML). The proposed method successfully improved precision 14, recall 22, and Dice score 8 percent compared to a binary cross-entropy loss function. Overall, the results of this work suggest that the proposed normalized active-passive loss using soft labels successfully mitigated the effects of noisy labels.

cs.CV↗

arXiv2022

Self-Supervised-RCNN for Medical Image Segmentation with Limited Data Annotation

Many successful methods developed for medical image analysis that are based on machine learning use supervised learning approaches, which often require large datasets annotated by experts to achieve high accuracy. However, medical data annotation is time-consuming and expensive, especially for segmentation tasks. To solve the problem of learning with limited labeled medical image data, an alternative deep learning training strategy based on self-supervised pretraining on unlabeled MRI scans is proposed in this work. Our pretraining approach first, randomly applies different distortions to random areas of unlabeled images and then predicts the type of distortions and loss of information. To this aim, an improved version of Mask-RCNN architecture has been adapted to localize the distortion location and recover the original image pixels. The effectiveness of the proposed method for segmentation tasks in different pre-training and fine-tuning scenarios is evaluated based on the Osteoarthritis Initiative dataset. Using this self-supervised pretraining method improved the Dice score by 20% compared to training from scratch. The proposed self-supervised learning is simple, effective, and suitable for different ranges of medical image analysis tasks including anomaly detection, segmentation, and classification.

cs.CV↗

arXiv2022

Design of a Head Coil for High Resolution Mouse Brain Perfusion Imaging using Magnetic Particle Imaging

Magnetic Particle Imaging (MPI) is a novel and versatile imaging modality developing towards human application. When up-scaling to human size, the sensitivity of the systems naturally drops as the coil sensitivity depends on the bore diameter. Thus, new methods to push the sensitivity limit further have to be investigated to cope for this loss. In this paper a dedicated surface coil for mice is developed, improving the sensitivity in cerebral imaging applications. Similar to MRI the developed surface coil improves the sensitivity due to the closer vicinity to the region of interest. With the developed surface coil presented in this work, it is possible to image tracer samples containing only 896 pg Fe and detect even small vessels and anatomical structures within a wild type mouse model. As current sensitivity measures require a tracer system a new method for determining a sensitivity measure without this requirement is presented and verified to enable comparison between MPI receiver systems.

physics.med-ph↗

arXiv2021

Bone Adaptation as a Geometric Flow

This paper presents bone adaptation as a geometric flow. The proposed method is based on two assumptions: first, that the bone surface is smooth (not fractal) permitting the definition of a tangent plane and, second, that the interface between marrow and bone tissue phases is orientable. This permits the analysis of bone adaptation using the well-developed mathematics of geometric flows and the numerical techniques of the level set method. Most importantly, topological changes such as holes forming in plates and rods disconnecting can be treated formally and simulated naturally. First, the relationship between biological theories of bone adaptation and the mathematical object describing geometric flow is described. This is termed the adaptation function, $F$, and is the multi-scale link described by Frost's Utah paradigm between cellular dynamics and bone structure. Second, a model of age-related bone loss termed curvature-based bone adaptation is presented. Using previous literature, it is shown that curvature-based bone adaptation is the limiting continuous equation of simulated bone atrophy, a discrete model of bone aging. Interestingly, the parameters of the model can be defined in such a way that the flow is volume-preserving. This implies that bone health can in principle change in ways that fundamentally cannot be measured by areal or volumetric bone mineral density, requiring structure-level imaging. Third, a numerical method is described and two in silico experiments are performed demonstrating the non-volume-preserving and volume-preserving cases. Taken together, recognition of bone adaptation as a geometric flow permits the recruitment of mathematical and numerical developments over the last 50 years to understanding and describing the complex surface of bone.

q-bio.QM↗

arXiv2021

Constructing High-Order Signed Distance Maps from Computed Tomography Data with Application to Bone Morphometry

An algorithm is presented for constructing high-order signed distance fields for two phase materials imaged with computed tomography. The signed distance field is high-order in that it is free of the quantization artifact associated with the distance transform of sampled signals. The narrowband is solved using a closest point algorithm extended for implicit embeddings that are not a signed distance field. The high-order fast sweeping algorithm is used to extend the narrowband to the remainder of the domain. The order of accuracy of the narrowband and extension methods are verified on ideal implicit surfaces. The method is applied to ten excised cubes of bovine trabecular bone. Localization of the surface, estimation of phase densities, and local morphometry is validated with these subjects. Since the embedding is high-order, gradients and thus curvatures can be accurately estimated locally in the image data.

eess.IV↗

arXiv2021

High-Order Signed Distance Transform of Sampled Signals

Signed distance transforms of sampled signals can be constructed better than the traditional exact signed distance transform. Such a transform is termed the high-order signed distance transform and is defined as satisfying three conditions: the Eikonal equation, recovery by a Heaviside function, and has an order of accuracy greater than unity away from the medial axis. Such a transform is an improvement to the classic notion of an exact signed distance transform because it does not exhibit artifacts of quantization. A large constant, linear time complexity high-order signed distance transform for arbitrary dimensionality sampled signals is developed based on the high order fast sweeping method. The transform is initialized with an exact signed distance transform and quantization corrected through an upwind solver for the boundary value Eikonal equation. The proposed method cannot attain arbitrary order of accuracy and is limited by the initialization method and non-uniqueness of the problem. However, meshed surfaces are visually smoother and do not exhibit artifacts of quantization in local mean and Gaussian curvature.

eess.IV↗

arXiv2021

Local Morphometry of Closed, Implicit Surfaces

Anatomical structures such as the hippocampus, liver, and bones can be analyzed as orientable, closed surfaces. This permits the computation of volume, surface area, mean curvature, Gaussian curvature, and the Euler-Poincaré characteristic as well as comparison of these morphometrics between structures of different topology. The structures are commonly represented implicitly in curve evolution problems as the zero level set of an embedding. Practically, binary images of anatomical structures are embedded using a signed distance transform. However, quantization prevents the accurate computation of curvatures, leading to considerable errors in morphometry. This paper presents a fast, simple embedding procedure for accurate local morphometry as the zero crossing of the Gaussian blurred binary image. The proposed method was validated based on the femur and fourth lumbar vertebrae of 50 clinical computed tomography datasets. The results show that the signed distance transform leads to large quantization errors in the computed local curvature. Global validation of morphometry using regression and Bland-Altman analysis revealed that the coefficient of determination for the average mean curvature is improved from 93.8% with the signed distance transform to 100% with the proposed method. For the surface area, the proportional bias is improved from -5.0% for the signed distance transform to +0.6% for the proposed method. The Euler-Poincaré characteristic is improved from unusable in the signed distance transform to 98% accuracy for the proposed method. The proposed method enables an improved local and global evaluation of curvature for purposes of morphometry on closed, implicit surfaces.

cs.CV↗

arXiv2020

Sphere Under Advection and Mean Curvature Flow

Advection and mean curvature flow is used as a model of bone microarchitecture adaptation. It is an equivalent geometric flow to prescribed mean curvature flow with an additional rate term. In order to validate numerical methods for simulating this flow and developing an inverse solver, a closed-form solution for advection and mean curvature flow of a sphere is derived.

math.DG↗

arXiv2020

Bidirectional Modeling and Analysis of Brain Aging with Normalizing Flows

Brain aging is a widely studied longitudinal process throughout which the brain undergoes considerable morphological changes and various machine learning approaches have been proposed to analyze it. Within this context, brain age prediction from structural MR images and age-specific brain morphology template generation are two problems that have attracted much attention. While most approaches tackle these tasks independently, we assume that they are inverse directions of the same functional bidirectional relationship between a brain's morphology and an age variable. In this paper, we propose to model this relationship with a single conditional normalizing flow, which unifies brain age prediction and age-conditioned generative modeling in a novel way. In an initial evaluation of this idea, we show that our normalizing flow brain aging model can accurately predict brain age while also being able to generate age-specific brain morphology templates that realistically represent the typical aging trend in a healthy population. This work is a step towards unified modeling of functional relationships between 3D brain morphology and clinical variables of interest with powerful normalizing flows.

cs.CV↗

arXiv2020

Artifacts of Quantization in Distance Transforms

Distance transforms are a central tool in shape analysis, morphometry, and curve evolution problems. This work describes and investigates an artifact present in distance maps computed from sampled signals. Namely, sampling reflects through the distance transform causing quantization in the resulting distance map. Gradients of the quantized distance map show banding, affecting the quality of subsequence processing. Furthermore, this error is independent of the sampling period of the signal and cannot be removed by modifying the number of samples across an objects width. Where needed, distance maps should be computed from representations other than binary images. In the case where exact representations are needed, a dithering and noise removal algorithm is proposed.

eess.IV↗