Andreas Maier

Thomas Wimmer, Vladimir Golkov, Hoai Nam Dang et al. · eth-zurich

The ability of convolutional neural networks (CNNs) to recognize objects regardless of their position in the image is due to the translation-equivariance of the convolutional operation. Group-equivariant CNNs transfer this equivariance to other transformations of the input. Dealing appropriately with objects and object parts of different scale is challenging, and scale can vary for multiple reasons such as the underlying object size or the resolution of the imaging modality. In this paper, we propose a scale-equivariant convolutional network layer for three-dimensional data that guarantees scale-equivariance in 3D CNNs. Scale-equivariance lifts the burden of having to learn each possible scale separately, allowing the neural network to focus on higher-level learning goals, which leads to better results and better data-efficiency. We provide an overview of the theoretical foundations and scientific work on scale-equivariant neural networks in the two-dimensional domain. We then transfer the concepts from 2D to the three-dimensional space and create a scale-equivariant convolutional layer for 3D data. Using the proposed scale-equivariant layer, we create a scale-equivariant U-Net for medical image segmentation and compare it with a non-scale-equivariant baseline method. Our experiments demonstrate the effectiveness of the proposed method in achieving scale-equivariance for 3D medical image analysis. We publish our code at https://github.com/wimmerth/scale-equivariant-3d-convnet for further research and application.

Hendrik Schröter, Alberto N. Escalante-B., Tobias Rosenkranz et al.

Deep learning-based speech enhancement has seen huge improvements and recently also expanded to full band audio (48 kHz). However, many approaches have a rather high computational complexity and require big temporal buffers for real time usage e.g. due to temporal convolutions or attention. Both make those approaches not feasible on embedded devices. This work further extends DeepFilterNet, which exploits harmonic structure of speech allowing for efficient speech enhancement (SE). Several optimizations in the training procedure, data augmentation, and network structure result in state-of-the-art SE performance while reducing the real-time factor to 0.04 on a notebook Core-i5 CPU. This makes the algorithm applicable to run on embedded devices in real-time. The DeepFilterNet framework can be obtained under an open source license.

Mingxuan Gu, Sulaiman Vesal, Mareike Thies et al. · stanford

Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a labeled source domain to an unlabeled target domain. Contrastive learning (CL) in the context of UDA can help to better separate classes in feature space. However, in image segmentation, the large memory footprint due to the computation of the pixel-wise contrastive loss makes it prohibitive to use. Furthermore, labeled target data is not easily available in medical imaging, and obtaining new samples is not economical. As a result, in this work, we tackle a more challenging UDA task when there are only a few (fewshot) or a single (oneshot) image available from the target domain. We apply a style transfer module to mitigate the scarcity of target samples. Then, to align the source and target features and tackle the memory issue of the traditional contrastive loss, we propose the centroid-based contrastive learning (CCL) and a centroid norm regularizer (CNR) to optimize the contrastive pairs in both direction and magnitude. In addition, we propose multi-partition centroid contrastive learning (MPCCL) to further reduce the variance in the target features. Fewshot evaluation on MS-CMRSeg dataset demonstrates that ConFUDA improves the segmentation performance by 0.34 of the Dice score on the target domain compared with the baseline, and 0.31 Dice score improvement in a more rigorous oneshot setting.

Marc Aubreville, Nikolas Stathonikos, Christof A. Bertram et al.

The density of mitotic figures within tumor tissue is known to be highly correlated with tumor proliferation and thus is an important marker in tumor grading. Recognition of mitotic figures by pathologists is known to be subject to a strong inter-rater bias, which limits the prognostic value. State-of-the-art deep learning methods can support the expert in this assessment but are known to strongly deteriorate when applied in a different clinical environment than was used for training. One decisive component in the underlying domain shift has been identified as the variability caused by using different whole slide scanners. The goal of the MICCAI MIDOG 2021 challenge has been to propose and evaluate methods that counter this domain shift and derive scanner-agnostic mitosis detection algorithms. The challenge used a training set of 200 cases, split across four scanning systems. As a test set, an additional 100 cases split across four scanning systems, including two previously unseen scanners, were given. The best approaches performed on an expert level, with the winning algorithm yielding an F_1 score of 0.748 (CI95: 0.704-0.781). In this paper, we evaluate and compare the approaches that were submitted to the challenge and identify methodological factors contributing to better performance.

Thomas Weissmann, Yixing Huang, Stefan Fischer et al.

Background: Deep learning (DL)-based head and neck lymph node level (HN_LNL) autodelineation is of high relevance to radiotherapy research and clinical treatment planning but still underinvestigated in academic literature. Methods: An expert-delineated cohort of 35 planning CTs was used for training of an nnU-net 3D-fullres/2D-ensemble model for autosegmentation of 20 different HN_LNL. A second cohort acquired at the same institution later in time served as the test set (n=20). In a completely blinded evaluation, 3 clinical experts rated the quality of DL autosegmentations in a head-to-head comparison with expert-created contours. For a subgroup of 10 cases, intraobserver variability was compared to the average DL autosegmentation accuracy on the original and recontoured set of expert segmentations. A postprocessing step to adjust craniocaudal boundaries of level autosegmentations to the CT slice plane was introduced and the effect on geometric accuracy and expert rating was investigated. Results: Blinded expert ratings for DL segmentations and expert-created contours were not significantly different. DL segmentations with slice plane adjustment were rated numerically higher (mean, 81.0 vs. 79.6,p=0.185) and DL segmentations without slice plane adjustment were rated numerically lower (77.2 vs. 79.6,p=0.167) than manually drawn contours. DL segmentations with CT slice plane adjustment were rated significantly better than DL contours without slice plane adjustment (81.0 vs. 77.2,p=0.004). Geometric accuracy of DL segmentations was not different from intraobserver variability (mean, 0.76 vs. 0.77, p=0.307). Conclusions: We show that a nnU-net 3D-fullres/2D-ensemble model can be used for highly accurate autodelineation of HN_LNL using only a limited training dataset that is ideally suited for large-scale standardized autodelineation of HN_LNL in the research setting.

Abner Hernandez, Paula Andrea Pérez-Toro, Elmar Nöth et al.

State-of-the-art automatic speech recognition (ASR) systems perform well on healthy speech. However, the performance on impaired speech still remains an issue. The current study explores the usefulness of using Wav2Vec self-supervised speech representations as features for training an ASR system for dysarthric speech. Dysarthric speech recognition is particularly difficult as several aspects of speech such as articulation, prosody and phonation can be impaired. Specifically, we train an acoustic model with features extracted from Wav2Vec, Hubert, and the cross-lingual XLSR model. Results suggest that speech representations pretrained on large unlabelled data can improve word error rate (WER) performance. In particular, features from the multilingual model led to lower WERs than filterbanks (Fbank) or models trained on a single language. Improvements were observed in English speakers with cerebral palsy caused dysarthria (UASpeech corpus), Spanish speakers with Parkinsonian dysarthria (PC-GITA corpus) and Italian speakers with paralysis-based dysarthria (EasyCall corpus). Compared to using Fbank features, XLSR-based features reduced WERs by 6.8%, 22.0%, and 7.0% for the UASpeech, PC-GITA, and EasyCall corpus, respectively.

Yixing Huang, Christoph Bert, Stefan Fischer et al.

Due to data privacy constraints, data sharing among multiple centers is restricted. Continual learning, as one approach to peer-to-peer federated learning, can promote multicenter collaboration on deep learning algorithm development by sharing intermediate models instead of training data. This work aims to investigate the feasibility of continual learning for multicenter collaboration on an exemplary application of brain metastasis identification using DeepMedic. 920 T1 MRI contrast enhanced volumes are split to simulate multicenter collaboration scenarios. A continual learning algorithm, synaptic intelligence (SI), is applied to preserve important model weights for training one center after another. In a bilateral collaboration scenario, continual learning with SI achieves a sensitivity of 0.917, and naive continual learning without SI achieves a sensitivity of 0.906, while two models trained on internal data solely without continual learning achieve sensitivity of 0.853 and 0.831 only. In a seven-center multilateral collaboration scenario, the models trained on internal datasets (100 volumes each center) without continual learning obtain a mean sensitivity value of 0.699. With single-visit continual learning (i.e., the shared model visits each center only once during training), the sensitivity is improved to 0.788 and 0.849 without SI and with SI, respectively. With iterative continual learning (i.e., the shared model revisits each center multiple times during training), the sensitivity is further improved to 0.914, which is identical to the sensitivity using mixed data for training. Our experiments demonstrate that continual learning can improve brain metastasis identification performance for centers with limited data. This study demonstrates the feasibility of applying continual learning for peer-to-peer federated learning in multicenter collaboration.

Mathias Öttl, Jana Mönius, Matthias Rübner et al.

Tumor segmentation in histopathology images is often complicated by its composition of different histological subtypes and class imbalance. Oversampling subtypes with low prevalence features is not a satisfactory solution since it eventually leads to overfitting. We propose to create synthetic images with semantically-conditioned deep generative networks and to combine subtype-balanced synthetic images with the original dataset to achieve better segmentation performance. We show the suitability of Generative Adversarial Networks (GANs) and especially diffusion models to create realistic images based on subtype-conditioning for the use case of HER2-stained histopathology. Additionally, we show the capability of diffusion models to conditionally inpaint HER2 tumor areas with modified subtypes. Combining the original dataset with the same amount of diffusion-generated images increased the tumor Dice score from 0.833 to 0.854 and almost halved the variance between the HER2 subtype recalls. These results create the basis for more reliable automatic HER2 analysis with lower performance variance between individual HER2 subtypes.

Yixing Huang, Ahmed Gomaa, Sabine Semrau et al.

The potential of large language models in medicine for education and decision making purposes has been demonstrated as they achieve decent scores on medical exams such as the United States Medical Licensing Exam (USMLE) and the MedQA exam. In this work, we evaluate the performance of ChatGPT-4 in the specialized field of radiation oncology using the 38th American College of Radiology (ACR) radiation oncology in-training (TXIT) exam and the 2022 Red Journal Gray Zone cases. For the TXIT exam, ChatGPT-3.5 and ChatGPT-4 have achieved the scores of 63.65% and 74.57%, respectively, highlighting the advantage of the latest ChatGPT-4 model. Based on the TXIT exam, ChatGPT-4's strong and weak areas in radiation oncology are identified to some extent. Specifically, ChatGPT-4 demonstrates better knowledge of statistics, CNS & eye, pediatrics, biology, and physics than knowledge of bone & soft tissue and gynecology, as per the ACR knowledge domain. Regarding clinical care paths, ChatGPT-4 performs better in diagnosis, prognosis, and toxicity than brachytherapy and dosimetry. It lacks proficiency in in-depth details of clinical trials. For the Gray Zone cases, ChatGPT-4 is able to suggest a personalized treatment approach to each case with high correctness and comprehensiveness. Importantly, it provides novel treatment aspects for many cases, which are not suggested by any human experts. Both evaluations demonstrate the potential of ChatGPT-4 in medical education for the general public and cancer patients, as well as the potential to aid clinical decision-making, while acknowledging its limitations in certain domains. Because of the risk of hallucination, facts provided by ChatGPT always need to be verified.

Karthik Shetty, Annette Birkhold, Srikrishna Jaganathan et al.

Objective: A digital twin of a patient can be a valuable tool for enhancing clinical tasks such as workflow automation, patient-specific X-ray dose optimization, markerless tracking, positioning, and navigation assistance in image-guided interventions. However, it is crucial that the patient's surface and internal organs are of high quality for any pose and shape estimates. At present, the majority of statistical shape models (SSMs) are restricted to a small number of organs or bones or do not adequately represent the general population. Method: To address this, we propose a deformable human shape and pose model that combines skin, internal organs, and bones, learned from CT images. By modeling the statistical variations in a pose-normalized space using probabilistic PCA while also preserving joint kinematics, our approach offers a holistic representation of the body that can benefit various medical applications. Results: We assessed our model's performance on a registered dataset, utilizing the unified shape space, and noted an average error of 3.6 mm for bones and 8.8 mm for organs. To further verify our findings, we conducted additional tests on publicly available datasets with multi-part segmentations, which confirmed the effectiveness of our model. Conclusion: This works shows that anatomically parameterized statistical shape models can be created accurately and in a computationally efficient manner. Significance: The proposed approach enables the construction of shape models that can be directly applied to various medical applications, including biomechanics and reconstruction.

Leonid Mill, Oliver Aust, Jochen A. Ackermann et al.

Artificial intelligence (AI), machine learning, and deep learning (DL) methods are becoming increasingly important in the field of biomedical image analysis. However, to exploit the full potential of such methods, a representative number of experimentally acquired images containing a significant number of manually annotated objects is needed as training data. Here we introduce SYNTA (synthetic data) as a novel approach for the generation of synthetic, photo-realistic, and highly complex biomedical images as training data for DL systems. We show the versatility of our approach in the context of muscle fiber and connective tissue analysis in histological sections. We demonstrate that it is possible to perform robust and expert-level segmentation tasks on previously unseen real-world data, without the need for manual annotations using synthetic training data alone. Being a fully parametric technique, our approach poses an interpretable and controllable alternative to Generative Adversarial Networks (GANs) and has the potential to significantly accelerate quantitative image analysis in a variety of biomedical applications in microscopy and beyond.

Siyuan Mei, Fuxin Fan, Andreas Maier

During orthopaedic surgery, the inserting of metallic implants or screws are often performed under mobile C-arm systems. Due to the high attenuation of metals, severe metal artifacts occur in 3D reconstructions, which degrade the image quality greatly. To reduce the artifacts, many metal artifact reduction algorithms have been developed and metal inpainting in projection domain is an essential step. In this work, a score-based generative model is trained on simulated knee projections and the inpainted image is obtained by removing the noise in conditional resampling process. The result implies that the inpainted images by score-based generative model have more detailed information and achieve the lowest mean absolute error and the highest peak-signal-to-noise-ratio compared with interpolation and CNN based method. Besides, the score-based model can also recover projections with big circlar and rectangular masks, showing its generalization in inpainting task.

Fei Wu, Nora Gourmelon, Thorsten Seehaus et al.

Knowledge on changes in glacier calving front positions is important for assessing the status of glaciers. Remote sensing imagery provides the ideal database for monitoring calving front positions, however, it is not feasible to perform this task manually for all calving glaciers globally due to time-constraints. Deep learning-based methods have shown great potential for glacier calving front delineation from optical and radar satellite imagery. The calving front is represented as a single thin line between the ocean and the glacier, which makes the task vulnerable to inaccurate predictions. The limited availability of annotated glacier imagery leads to a lack of data diversity (not all possible combinations of different weather conditions, terminus shapes, sensors, etc. are present in the data), which exacerbates the difficulty of accurate segmentation. In this paper, we propose Attention-Multi-hooking-Deep-supervision HookNet (AMD-HookNet), a novel glacier calving front segmentation framework for synthetic aperture radar (SAR) images. The proposed method aims to enhance the feature representation capability through multiple information interactions between low-resolution and high-resolution inputs based on a two-branch U-Net. The attention mechanism, integrated into the two branch U-Net, aims to interact between the corresponding coarse and fine-grained feature maps. This allows the network to automatically adjust feature relationships, resulting in accurate pixel-classification predictions. Extensive experiments and comparisons on the challenging glacier segmentation benchmark dataset CaFFe show that our AMD-HookNet achieves a mean distance error of 438 m to the ground truth outperforming the current state of the art by 42%, which validates its effectiveness.

Kubilay Can Demir, Matthias May, Axel Schmid et al.

This paper presents a new multimodal interventional radiology dataset, called PoCaP (Port Catheter Placement) Corpus. This corpus consists of speech and audio signals in German, X-ray images, and system commands collected from 31 PoCaP interventions by six surgeons with average duration of 81.4 $\pm$ 41.0 minutes. The corpus aims to provide a resource for developing a smart speech assistant in operating rooms. In particular, it may be used to develop a speech controlled system that enables surgeons to control the operation parameters such as C-arm movements and table positions. In order to record the dataset, we acquired consent by the institutional review board and workers council in the University Hospital Erlangen and by the patients for data privacy. We describe the recording set-up, data structure, workflow and preprocessing steps, and report the first PoCaP Corpus speech recognition analysis results with 11.52 $\%$ word error rate using pretrained models. The findings suggest that the data has the potential to build a robust command recognition system and will allow the development of a novel intervention support systems using speech and image processing in the medical domain.

Karthik Shetty, Annette Birkhold, Srikrishna Jaganathan et al.

We introduce PLIKS (Pseudo-Linear Inverse Kinematic Solver) for reconstruction of a 3D mesh of the human body from a single 2D image. Current techniques directly regress the shape, pose, and translation of a parametric model from an input image through a non-linear mapping with minimal flexibility to any external influences. We approach the task as a model-in-the-loop optimization problem. PLIKS is built on a linearized formulation of the parametric SMPL model. Using PLIKS, we can analytically reconstruct the human model via 2D pixel-aligned vertices. This enables us with the flexibility to use accurate camera calibration information when available. PLIKS offers an easy way to introduce additional constraints such as shape and translation. We present quantitative evaluations which confirm that PLIKS achieves more accurate reconstruction with greater than 10% improvement compared to other state-of-the-art methods with respect to the standard 3D human pose and shape benchmarks while also obtaining a reconstruction error improvement of 12.9 mm on the newer AGORA dataset.

Md Hasan, Nyvenn Castro, Daiqi Liu et al.

Real-time magnetic resonance imaging (rtMRI) of speech production enables non-invasive visualization of dynamic vocal-tract motion and is valuable for speech science and clinical assessment. However, rtMRI is fundamentally constrained by trade-offs among spatial resolution, temporal resolution, and acquisition speed, often leading to undersampled k-space measurements and degraded reconstructions. We propose SIREM, a speech-informed MRI reconstruction framework that uses synchronized speech as a cross-modal prior. The central idea is that vocal-tract configurations during speech are correlated with the produced acoustics, making part of the image content predictable from audio. SIREM models each frame as a fusion of an audio-driven component and an MRI-driven component through a spatial weighting map. The audio branch predicts articulator-related structure from speech, while the MRI branch reconstructs complementary content from measured k-space data. We further introduce a learnable soft weighting profile over spiral arms, enabling a differentiable study of how k-space arm usage interacts with speech-informed fusion. This yields a unified multimodal formulation that combines audio-driven prediction, MRI reconstruction, and sampling adaptation. We evaluate SIREM on the USC speech rtMRI benchmark against standard baselines, including gridding, wavelet-based compressed sensing, and total variation. SIREM introduces a speech-informed reconstruction paradigm that operates in a substantially higher-throughput regime than iterative methods while preserving anatomically plausible vocal-tract structure. These results establish an initial benchmark for multimodal speech-informed rtMRI reconstruction and highlight the potential of synchronized speech as an auxiliary prior for fast reconstruction. The source code is available at https://github.com/mdhasanai/SIREM

Mayank Patwari, Ralf Gutjahr, Rainer Raupach et al.

The use of deep learning has successfully solved several problems in the field of medical imaging. Deep learning has been applied to the CT denoising problem successfully. However, the use of deep learning requires large amounts of data to train deep convolutional networks (CNNs). Moreover, due to large parameter count, such deep CNNs may cause unexpected results. In this study, we introduce a novel CT denoising framework, which has interpretable behaviour, and provides useful results with limited data. We employ bilateral filtering in both the projection and volume domains to remove noise. To account for non-stationary noise, we tune the $σ$ parameters of the volume for every projection view, and for every volume pixel. The tuning is carried out by two deep CNNs. Due to impracticality of labelling, the two deep CNNs are trained via a Deep-Q reinforcement learning task. The reward for the task is generated by using a custom reward function represented by a neural network. Our experiments were carried out on abdominal scans for the Mayo Clinic TCIA dataset, and the AAPM Low Dose CT Grand Challenge. Our denoising framework has excellent denoising performance increasing the PSNR from 28.53 to 28.93, and increasing the SSIM from 0.8952 to 0.9204. We outperform several state-of-the-art deep CNNs, which have several orders of magnitude higher number of parameters (p-value (PSNR) = 0.000, p-value (SSIM) = 0.000). Our method does not introduce any blurring, which is introduced by MSE loss based methods, or any deep learning artifacts, which are introduced by WGAN based models. Our ablation studies show that parameter tuning and using our reward network results in the best possible results.

Yihao Hou, Christoph Bert, Ahmed Gomaa et al.

Generating physician letters is a time-consuming task in daily clinical practice. This study investigates local fine-tuning of large language models (LLMs), specifically LLaMA models, for physician letter generation in a privacy-preserving manner within the field of radiation oncology. Our findings demonstrate that base LLaMA models, without fine-tuning, are inadequate for effectively generating physician letters. The QLoRA algorithm provides an efficient method for local intra-institutional fine-tuning of LLMs with limited computational resources (i.e., a single 48 GB GPU workstation within the hospital). The fine-tuned LLM successfully learns radiation oncology-specific information and generates physician letters in an institution-specific style. ROUGE scores of the generated summary reports highlight the superiority of the 8B LLaMA-3 model over the 13B LLaMA-2 model. Further multidimensional physician evaluations of 10 cases reveal that, although the fine-tuned LLaMA-3 model has limited capacity to generate content beyond the provided input data, it successfully generates salutations, diagnoses and treatment histories, recommendations for further treatment, and planned schedules. Overall, clinical benefit was rated highly by the clinical experts (average score of 3.44 on a 4-point scale). With careful physician review and correction, automated LLM-based physician letter generation has significant practical value.

Marcel Reimann, Jungeun Won, Hiroyuki Takahashi et al.

Recent advances in optical coherence tomography such as the development of high speed ultrahigh resolution scanners and corresponding signal processing techniques may reveal new potential biomarkers in retinal diseases. Newly visible features are, for example, small hyperreflective specks in age-related macular degeneration. Identifying these new markers is crucial to investigate potential association with disease progression and treatment outcomes. Therefore, it is necessary to reliably detect these features in 3D volumetric scans. Because manual labeling of entire volumes is infeasible a need for automatic detection arises. Labeled datasets are often not publicly available and there are usually large variations in scan protocols and scanner types. Thus, this work focuses on an unsupervised approach that is based on local peak-detection and random walker segmentation to detect small features on each B-scan of the volume.

Frauke Wilm, Marco Fragoso, Christof A. Bertram et al.

In histopathology, scanner-induced domain shifts are known to impede the performance of trained neural networks when tested on unseen data. Multi-domain pre-training or dedicated domain-generalization techniques can help to develop domain-agnostic algorithms. For this, multi-scanner datasets with a high variety of slide scanning systems are highly desirable. We present a publicly available multi-scanner dataset of canine cutaneous squamous cell carcinoma histopathology images, composed of 44 samples digitized with five slide scanners. This dataset provides local correspondences between images and thereby isolates the scanner-induced domain shift from other inherent, e.g. morphology-induced domain shifts. To highlight scanner differences, we present a detailed evaluation of color distributions, sharpness, and contrast of the individual scanner subsets. Additionally, to quantify the inherent scanner-induced domain shift, we train a tumor segmentation network on each scanner subset and evaluate the performance both in- and cross-domain. We achieve a class-averaged in-domain intersection over union coefficient of up to 0.86 and observe a cross-domain performance decrease of up to 0.38, which confirms the inherent domain shift of the presented dataset and its negative impact on the performance of deep neural networks.

Mareike Thies, Fabian Wagner, Noah Maul et al.

Incorporating computed tomography (CT) reconstruction operators into differentiable pipelines has proven beneficial in many applications. Such approaches usually focus on the projection data and keep the acquisition geometry fixed. However, precise knowledge of the acquisition geometry is essential for high quality reconstruction results. In this paper, the differentiable formulation of fan-beam CT reconstruction is extended to the acquisition geometry. This allows to propagate gradient information from a loss function on the reconstructed image into the geometry parameters. As a proof-of-concept experiment, this idea is applied to rigid motion compensation. The cost function is parameterized by a trained neural network which regresses an image quality metric from the motion affected reconstruction alone. Using the proposed method, we are the first to optimize such an autofocus-inspired algorithm based on analytical gradients. The algorithm achieves a reduction in MSE by 35.5 % and an improvement in SSIM by 12.6 % over the motion affected reconstruction. Next to motion compensation, we see further use cases of our differentiable method for scanner calibration or hybrid techniques employing deep models.

Frauke Wilm, Marco Fragoso, Christof A. Bertram et al.

Computer-aided systems in histopathology are often challenged by various sources of domain shift that impact the performance of these algorithms considerably. We investigated the potential of using self-supervised pre-training to overcome scanner-induced domain shifts for the downstream task of tumor segmentation. For this, we present the Barlow Triplets to learn scanner-invariant representations from a multi-scanner dataset with local image correspondences. We show that self-supervised pre-training successfully aligned different scanner representations, which, interestingly only results in a limited benefit for our downstream task. We thereby provide insights into the influence of scanner characteristics for downstream applications and contribute to a better understanding of why established self-supervised methods have not yet shown the same success on histopathology data as they have for natural images.

Soroosh Tayebi Arasteh, Tobias Weise, Maria Schuster et al.

Navigating the challenges of data-driven speech processing, one of the primary hurdles is accessing reliable pathological speech data. While public datasets appear to offer solutions, they come with inherent risks of potential unintended exposure of patient health information via re-identification attacks. Using a comprehensive real-world pathological speech corpus, with over n=3,800 test subjects spanning various age groups and speech disorders, we employed a deep-learning-driven automatic speaker verification (ASV) approach. This resulted in a notable mean equal error rate (EER) of 0.89% with a standard deviation of 0.06%, outstripping traditional benchmarks. Our comprehensive assessments demonstrate that pathological speech overall faces heightened privacy breach risks compared to healthy speech. Specifically, adults with dysphonia are at heightened re-identification risks, whereas conditions like dysarthria yield results comparable to those of healthy speakers. Crucially, speech intelligibility does not influence the ASV system's performance metrics. In pediatric cases, particularly those with cleft lip and palate, the recording environment plays a decisive role in re-identification. Merging data across pathological types led to a marked EER decrease, suggesting the potential benefits of pathological diversity in ASV, accompanied by a logarithmic boost in ASV effectiveness. In essence, this research sheds light on the dynamics between pathological speech and speaker verification, emphasizing its crucial role in safeguarding patient confidentiality in our increasingly digitized healthcare era.

Susu Sun, Stefano Woerner, Andreas Maier et al.

Interpretability is essential for machine learning algorithms in high-stakes application fields such as medical image analysis. However, high-performing black-box neural networks do not provide explanations for their predictions, which can lead to mistrust and suboptimal human-ML collaboration. Post-hoc explanation techniques, which are widely used in practice, have been shown to suffer from severe conceptual problems. Furthermore, as we show in this paper, current explanation techniques do not perform adequately in the multi-label scenario, in which multiple medical findings may co-occur in a single image. We propose Attri-Net, an inherently interpretable model for multi-label classification. Attri-Net is a powerful classifier that provides transparent, trustworthy, and human-understandable explanations. The model first generates class-specific attribution maps based on counterfactuals to identify which image regions correspond to certain medical findings. Then a simple logistic regression classifier is used to make predictions based solely on these attribution maps. We compare Attri-Net to five post-hoc explanation techniques and one inherently interpretable classifier on three chest X-ray datasets. We find that Attri-Net produces high-quality multi-label explanations consistent with clinical knowledge and has comparable classification performance to state-of-the-art classification models.

Paul Stoewer, Achim Schilling, Andreas Maier et al.

The hippocampal-entorhinal complex plays a major role in the organization of memory and thought. The formation of and navigation in cognitive maps of arbitrary mental spaces via place and grid cells can serve as a representation of memories and experiences and their relations to each other. The multi-scale successor representation is proposed to be the mathematical principle underlying place and grid cell computations. Here, we present a neural network, which learns a cognitive map of a semantic space based on 32 different animal species encoded as feature vectors. The neural network successfully learns the similarities between different animal species, and constructs a cognitive map of 'animal space' based on the principle of successor representations with an accuracy of around 30% which is near to the theoretical maximum regarding the fact that all animal species have more than one possible successor, i.e. nearest neighbor in feature space. Furthermore, a hierarchical structure, i.e. different scales of cognitive maps, can be modeled based on multi-scale successor representations. We find that, in fine-grained cognitive maps, the animal vectors are evenly distributed in feature space. In contrast, in coarse-grained maps, animal vectors are highly clustered according to their biological class, i.e. amphibians, mammals and insects. This could be a possible mechanism explaining the emergence of new abstract semantic concepts. Finally, even completely new or incomplete input can be represented by interpolation of the representations from the cognitive map with remarkable high accuracy of up to 95%. We conclude that the successor representation can serve as a weighted pointer to past memories and experiences, and may therefore be a crucial building block for future machine learning to include prior knowledge, and to derive context knowledge from novel input.

Mathias Zinnen, Prathmesh Madhu, Ronak Kosti et al.

The Odeuropa Challenge on Olfactory Object Recognition aims to foster the development of object detection in the visual arts and to promote an olfactory perspective on digital heritage. Object detection in historical artworks is particularly challenging due to varying styles and artistic periods. Moreover, the task is complicated due to the particularity and historical variance of predefined target objects, which exhibit a large intra-class variance, and the long tail distribution of the dataset labels, with some objects having only very few training examples. These challenges should encourage participants to create innovative approaches using domain adaptation or few-shot learning. We provide a dataset of 2647 artworks annotated with 20 120 tightly fit bounding boxes that are split into a training and validation set (public). A test set containing 1140 artworks and 15 480 annotations is kept private for the challenge evaluation.

Nico Meyer, Christopher Mutschler, Dominik Seuß et al.

Logical operations are essential for quantum computation within quantum error-correcting codes. However, discovering their physical realizations is challenging, especially for non-additive codes that lack a stabilizer description. We present a general learning-based framework that, given only an encoding circuit, constructs physical implementations of logical operations while enforcing structural properties such as transversality or shallow depth. Our approach is validated by rediscovering known logical operations of standard stabilizer codes. We then extend it to a co-design procedure, dubbed variational early fault-tolerant quantum computing (VarEFTQC), which tailors non-additive encodings to a given noise model and enforces desired logical gate sets, such as transversal IQP-type families or low-depth universal sets. A software library implements the complete learning pipeline, including loss-function variants, ansatz families, and optimization routines. Together, these results position VarEFTQC as a practical tool for discovering hardware-adapted logical gadgets for early fault-tolerant quantum computing.

Aline Sindel, Bettina Hohberger, Andreas Maier et al.

In ophthalmological imaging, multiple imaging systems, such as color fundus, infrared, fluorescein angiography, optical coherence tomography (OCT) or OCT angiography, are often involved to make a diagnosis of retinal disease. Multi-modal retinal registration techniques can assist ophthalmologists by providing a pixel-based comparison of aligned vessel structures in images from different modalities or acquisition times. To this end, we propose an end-to-end trainable deep learning method for multi-modal retinal image registration. Our method extracts convolutional features from the vessel structure for keypoint detection and description and uses a graph neural network for feature matching. The keypoint detection and description network and graph neural network are jointly trained in a self-supervised manner using synthetic multi-modal image pairs and are guided by synthetically sampled ground truth homographies. Our method demonstrates higher registration accuracy as competing methods for our synthetic retinal dataset and generalizes well for our real macula dataset and a public fundus dataset.

Yuan Lin, Murong Xu, Marc Hölle et al.

Widely adopted medical image segmentation methods, although efficient, are primarily deterministic and remain poorly amenable to natural language prompts. Thus, they lack the capability to estimate multiple proposals, human interaction, and cross-modality adaptation. Recently, text-to-image diffusion models have shown potential to bridge the gap. However, training them from scratch requires a large dataset-a limitation for medical image segmentation. Furthermore, they are often limited to binary segmentation and cannot be conditioned on a natural language prompt. To this end, we propose a novel framework called ProGiDiff that leverages existing image generation models for medical image segmentation purposes. Specifically, we propose a ControlNet-style conditioning mechanism with a custom encoder, suitable for image conditioning, to steer a pre-trained diffusion model to output segmentation masks. It naturally extends to a multi-class setting simply by prompting the target organ. Our experiment on organ segmentation from CT images demonstrates strong performance compared to previous methods and could greatly benefit from an expert-in-the-loop setting to leverage multiple proposals. Importantly, we demonstrate that the learned conditioning mechanism can be easily transferred through low-rank, few-shot adaptation to segment MR images.

Mathias Zinnen, Azhar Hussian, Hang Tran et al.

Smell gestures play a crucial role in the investigation of past smells in the visual arts yet their automated recognition poses significant challenges. This paper introduces the SniffyArt dataset, consisting of 1941 individuals represented in 441 historical artworks. Each person is annotated with a tightly fitting bounding box, 17 pose keypoints, and a gesture label. By integrating these annotations, the dataset enables the development of hybrid classification approaches for smell gesture recognition. The datasets high-quality human pose estimation keypoints are achieved through the merging of five separate sets of keypoint annotations per person. The paper also presents a baseline analysis, evaluating the performance of representative algorithms for detection, keypoint estimation, and classification tasks, showcasing the potential of combining keypoint estimation with smell gesture classification. The SniffyArt dataset lays a solid foundation for future research and the exploration of multi-task approaches leveraging pose keypoints and person boxes to advance human gesture and olfactory dimension analysis in historical artworks.

Achim Schilling, William Sedley, Richard Gerum et al.

How is information processed in the brain during perception? Mechanistic insight is achieved only when experiments are employed to test formal or computational models. In analogy to lesion studies, phantom perception may serve as a vehicle to understand the fundamental processing principles underlying auditory perception. With a special focus on tinnitus -- as the prime example of auditory phantom perception -- we review recent work at the intersection of artificial intelligence, psychology, and neuroscience. In particular, we discuss why everyone with tinnitus suffers from hearing loss, but not everyone with hearing loss suffers from tinnitus. We argue that the increase of sensory precision due to Bayesian inference could be caused by intrinsic neural noise and lead to a prediction error in the cerebral cortex. Hence, two fundamental processing principles - being ubiquitous in the brain - provide the most explanatory power for the emergence of tinnitus: predictive coding as a top-down, and stochastic resonance as a complementary bottom-up mechanism. We conclude that both principles play a crucial role in healthy auditory perception.

Aline Sindel, Andreas Maier, Vincent Christlein

Facial landmark detection plays an important role for the similarity analysis in artworks to compare portraits of the same or similar artists. With facial landmarks, portraits of different genres, such as paintings and prints, can be automatically aligned using control-point-based image registration. We propose a deep-learning-based method for facial landmark detection in high-resolution images of paintings and prints. It divides the task into a global network for coarse landmark prediction and multiple region networks for precise landmark refinement in regions of the eyes, nose, and mouth that are automatically determined based on the predicted global landmark coordinates. We created a synthetically augmented facial landmark art dataset including artistic style transfer and geometric landmark shifts. Our method demonstrates an accurate detection of the inner facial landmarks for our high-resolution dataset of artworks while being comparable for a public low-resolution artwork dataset in comparison to competing methods.

Mohammad Joshaghani, Amirabbas Davari, Faezeh Nejati Hatamian et al.

Employing deep neural networks for Hyperspectral remote sensing (HSRS) image classification is a challenging task. HSRS images have high dimensionality and a large number of channels with substantial redundancy between channels. In addition, the training data for classifying HSRS images is limited and the amount of available training data is much smaller compared to other classification tasks. These factors complicate the training process of deep neural networks with many parameters and cause them to not perform well even compared to conventional models. Moreover, convolutional neural networks produce over-confident predictions, which is highly undesirable considering the aforementioned problem. In this work, we use for HSRS image classification a special class of deep neural networks, namely a Bayesian neural network (BNN). To the extent of our knowledge, this is the first time that BNNs are used in HSRS image classification. BNNs inherently provide a measure for uncertainty. We perform extensive experiments on the Pavia Centre, Salinas, and Botswana datasets. We show that a BNN outperforms a standard convolutional neural network (CNN) and an off-the-shelf Random Forest (RF). Further experiments underline that the BNN is more stable and robust to model pruning, and that the uncertainty is higher for samples with higher expected prediction error.

Noah Maul, Katharina Zinn, Fabian Wagner et al.

Patient-specific hemodynamics assessment could support diagnosis and treatment of neurovascular diseases. Currently, conventional medical imaging modalities are not able to accurately acquire high-resolution hemodynamic information that would be required to assess complex neurovascular pathologies. Therefore, computational fluid dynamics (CFD) simulations can be applied to tomographic reconstructions to obtain clinically relevant information. However, three-dimensional (3D) CFD simulations require enormous computational resources and simulation-related expert knowledge that are usually not available in clinical environments. Recently, deep-learning-based methods have been proposed as CFD surrogates to improve computational efficiency. Nevertheless, the prediction of high-resolution transient CFD simulations for complex vascular geometries poses a challenge to conventional deep learning models. In this work, we present an architecture that is tailored to predict high-resolution (spatial and temporal) velocity fields for complex synthetic vascular geometries. For this, an octree-based spatial discretization is combined with an implicit neural function representation to efficiently handle the prediction of the 3D velocity field for each time step. The presented method is evaluated for the task of cerebral hemodynamics prediction before and during the injection of contrast agent in the internal carotid artery (ICA). Compared to CFD simulations, the velocity field can be estimated with a mean absolute error of 0.024 m/s, whereas the run time reduces from several hours on a high-performance cluster to a few seconds on a consumer graphical processing unit.

Mareike Thies, Fabian Wagner, Noah Maul et al.

In computed tomography (CT), the projection geometry used for data acquisition needs to be known precisely to obtain a clear reconstructed image. Rigid patient motion is a cause for misalignment between measured data and employed geometry. Commonly, such motion is compensated by solving an optimization problem that, e.g., maximizes the quality of the reconstructed image with respect to the projection geometry. So far, gradient-free optimization algorithms have been utilized to find the solution for this problem. Here, we show that gradient-based optimization algorithms are a possible alternative and compare the performance to their gradient-free counterparts on a benchmark motion compensation problem. Gradient-based algorithms converge substantially faster while being comparable to gradient-free algorithms in terms of capture range and robustness to the number of free parameters. Hence, gradient-based optimization is a viable alternative for the given type of problems.

Fuxin Fan, Ludwig Ritschl, Marcel Beister et al.

In several image acquisition and processing steps of X-ray radiography, knowledge of the existence of metal implants and their exact position is highly beneficial (e.g. dose regulation, image contrast adjustment). Another application which would benefit from an accurate metal segmentation is cone beam computed tomography (CBCT) which is based on 2D X-ray projections. Due to the high attenuation of metals, severe artifacts occur in the 3D X-ray acquisitions. The metal segmentation in CBCT projections usually serves as a prerequisite for metal artifact avoidance and reduction algorithms. Since the generation of high quality clinical training is a constant challenge, this study proposes to generate simulated X-ray images based on CT data sets combined with self-designed computer aided design (CAD) implants and make use of convolutional neural network (CNN) and vision transformer (ViT) for metal segmentation. Model test is performed on accurately labeled X-ray test datasets obtained from specimen scans. The CNN encoder-based network like U-Net has limited performance on cadaver test data with an average dice score below 0.30, while the metal segmentation transformer with dual decoder (MST-DD) shows high robustness and generalization on the segmentation task, with an average dice score of 0.90. Our study indicates that the CAD model-based data generation has high flexibility and could be a way to overcome the problem of shortage in clinical data sampling and labelling. Furthermore, the MST-DD approach generates a more reliable neural network in case of training on simulated data.

Amr Hagag, Ahmed Gomaa, Dominik Kornek et al.

Survival prediction for cancer patients is critical for optimal treatment selection and patient management. Current patient survival prediction methods typically extract survival information from patients' clinical record data or biological and imaging data. In practice, experienced clinicians can have a preliminary assessment of patients' health status based on patients' observable physical appearances, which are mainly facial features. However, such assessment is highly subjective. In this work, the efficacy of objectively capturing and using prognostic information contained in conventional portrait photographs using deep learning for survival predication purposes is investigated for the first time. A pre-trained StyleGAN2 model is fine-tuned on a custom dataset of our cancer patients' photos to empower its generator with generative ability suitable for patients' photos. The StyleGAN2 is then used to embed the photographs to its highly expressive latent space. Utilizing the state-of-the-art survival analysis models and based on StyleGAN's latent space photo embeddings, this approach achieved a C-index of 0.677, which is notably higher than chance and evidencing the prognostic value embedded in simple 2D facial images. In addition, thanks to StyleGAN's interpretable latent space, our survival prediction model can be validated for relying on essential facial features, eliminating any biases from extraneous information like clothing or background. Moreover, a health attribute is obtained from regression coefficients, which has important potential value for patient care.

Kishore Surendra, Achim Schilling, Paul Stoewer et al.

How do humans learn language, and can the first language be learned at all? These fundamental questions are still hotly debated. In contemporary linguistics, there are two major schools of thought that give completely opposite answers. According to Chomsky's theory of universal grammar, language cannot be learned because children are not exposed to sufficient data in their linguistic environment. In contrast, usage-based models of language assume a profound relationship between language structure and language use. In particular, contextual mental processing and mental representations are assumed to have the cognitive capacity to capture the complexity of actual language use at all levels. The prime example is syntax, i.e., the rules by which words are assembled into larger units such as sentences. Typically, syntactic rules are expressed as sequences of word classes. However, it remains unclear whether word classes are innate, as implied by universal grammar, or whether they emerge during language acquisition, as suggested by usage-based approaches. Here, we address this issue from a machine learning and natural language processing perspective. In particular, we trained an artificial deep neural network on predicting the next word, provided sequences of consecutive words as input. Subsequently, we analyzed the emerging activation patterns in the hidden layers of the neural network. Strikingly, we find that the internal representations of nine-word input sequences cluster according to the word class of the tenth word to be predicted as output, even though the neural network did not receive any explicit information about syntactic rules or word classes during training. This surprising result suggests, that also in the human brain, abstract representational categories such as word classes may naturally emerge as a consequence of predictive coding and processing during language acquisition.

Kai Packhäuser, Lukas Folle, Florian Thamm et al.

The availability of large-scale chest X-ray datasets is a requirement for developing well-performing deep learning-based algorithms in thoracic abnormality detection and classification. However, biometric identifiers in chest radiographs hinder the public sharing of such data for research purposes due to the risk of patient re-identification. To counteract this issue, synthetic data generation offers a solution for anonymizing medical images. This work employs a latent diffusion model to synthesize an anonymous chest X-ray dataset of high-quality class-conditional images. We propose a privacy-enhancing sampling strategy to ensure the non-transference of biometric information during the image generation process. The quality of the generated images and the feasibility of serving as exclusive training data are evaluated on a thoracic abnormality classification task. Compared to a real classifier, we achieve competitive results with a performance gap of only 3.5% in the area under the receiver operating characteristic curve.

Soroosh Tayebi Arasteh, Mahshad Lotfinia, Paula Andrea Perez-Toro et al.

Speech pathology has impacts on communication abilities and quality of life. While deep learning-based models have shown potential in diagnosing these disorders, the use of sensitive data raises critical privacy concerns. Although differential privacy (DP) has been explored in the medical imaging domain, its application in pathological speech analysis remains largely unexplored despite the equally critical privacy concerns. To the best of our knowledge, this study is the first to investigate DP's impact on pathological speech data, focusing on the trade-offs between privacy, diagnostic accuracy, and fairness. Using a large, real-world dataset of 200 hours of recordings from 2,839 German-speaking participants, we observed a maximum accuracy reduction of 3.85% when training with DP with high privacy levels. To highlight real-world privacy risks, we demonstrated the vulnerability of non-private models to gradient inversion attacks, reconstructing identifiable speech samples and showcasing DP's effectiveness in mitigating these risks. To explore the potential generalizability across languages and disorders, we validated our approach on a dataset of Spanish-speaking Parkinson's disease patients, leveraging pretrained models from healthy English-speaking datasets, and demonstrated that careful pretraining on large-scale task-specific datasets can maintain favorable accuracy under DP constraints. A comprehensive fairness analysis revealed minimal gender bias at reasonable privacy levels but underscored the need for addressing age-related disparities. Our results establish that DP can balance privacy and utility in speech disorder detection, while highlighting unique challenges in privacy-fairness trade-offs for speech data. This provides a foundation for refining DP methodologies and improving fairness across diverse patient groups in real-world deployments.

Mathias Zinnen, Prathmesh Madhu, Peter Bell et al.

We investigate the effect of style and category similarity in multiple datasets used for object detection pretraining. We find that including an additional stage of object-detection pretraining can increase the detection performance considerably. While our experiments suggest that style similarities between pre-training and target datasets are less important than matching categories, further experiments are needed to verify this hypothesis.

Philipp Sommer, Yixing Huang, Christoph Bert et al.

Stereotactic radiotherapy (SRT) is one of the most important treatment for patients with brain metastases (BM). Conventionally, following SRT patients are monitored by serial imaging and receive salvage treatments in case of significant tumor growth. We hypothesized that using radiomics and machine learning (ML), metastases at high risk for subsequent progression could be identified during follow-up prior to the onset of significant tumor growth, enabling personalized follow-up intervals and early selection for salvage treatment. All experiments are performed on a dataset from clinical routine of the Radiation Oncology department of the University Hospital Erlangen (UKER). The classification is realized via the maximum-relevance minimal-redundancy (MRMR) technique and support vector machines (SVM). The pipeline leads to a classification with a mean area under the curve (AUC) score of 0.83 in internal cross-validation and allows a division of the cohort into two subcohorts that differ significantly in their median time to progression (low-risk metastasis (LRM): 17.3 months, high-risk metastasis (HRM): 9.6 months, p < 0.01). The classification performance is especially enhanced by the analysis of medical images from different points in time (AUC 0.53 -> AUC 0.74). The results indicate that risk stratification of BM based on radiomics and machine learning during post-SRT follow-up is possible with good accuracy and should be further pursued to personalize and improve post-SRT follow-up.

Linda-Sophie Schneider, Mareike Thies, Christopher Syben et al.

We present a method for selecting valuable projections in computed tomography (CT) scans to enhance image reconstruction and diagnosis. The approach integrates two important factors, projection-based detectability and data completeness, into a single feed-forward neural network. The network evaluates the value of projections, processes them through a differentiable ranking function and makes the final selection using a straight-through estimator. Data completeness is ensured through the label provided during training. The approach eliminates the need for heuristically enforcing data completeness, which may exclude valuable projections. The method is evaluated on simulated data in a non-destructive testing scenario, where the aim is to maximize the reconstruction quality within a specified region of interest. We achieve comparable results to previous methods, laying the foundation for using reconstruction-based loss functions to learn the selection of projections.

Martin Mayr, Marcel Dreier, Florian Kordon et al.

The imitation of cursive handwriting is mainly limited to generating handwritten words or lines. Multiple synthetic outputs must be stitched together to create paragraphs or whole pages, whereby consistency and layout information are lost. To close this gap, we propose a method for imitating handwriting at the paragraph level that also works for unseen writing styles. Therefore, we introduce a modified latent diffusion model that enriches the encoder-decoder mechanism with specialized loss functions that explicitly preserve the style and content. We enhance the attention mechanism of the diffusion model with adaptive 2D positional encoding and the conditioning mechanism to work with two modalities simultaneously: a style image and the target text. This significantly improves the realism of the generated handwriting. Our approach sets a new benchmark in our comprehensive evaluation. It outperforms all existing imitation methods at both line and paragraph levels, considering combined style and content preservation.

Tobias Weise, Philipp Klumpp, Kubilay Can Demir et al.

Speech intelligibility assessment plays an important role in the therapy of patients suffering from pathological speech disorders. Automatic and objective measures are desirable to assist therapists in their traditionally subjective and labor-intensive assessments. In this work, we investigate a novel approach for obtaining such a measure using the divergence in disentangled latent speech representations of a parallel utterance pair, obtained from a healthy reference and a pathological speaker. Experiments on an English database of Cerebral Palsy patients, using all available utterances per speaker, show high and significant correlation values (R = -0.9) with subjective intelligibility measures, while having only minimal deviation (+-0.01) across four different reference speaker pairs. We also demonstrate the robustness of the proposed method (R = -0.89 deviating +-0.02 over 1000 iterations) by considering a significantly smaller amount of utterances per speaker. Our results are among the first to show that disentangled speech representations can be used for automatic pathological speech intelligibility assessment, resulting in a reference speaker pair invariant method, applicable in scenarios with only few utterances available.

Paul Stoewer, Achim Schilling, Andreas Maier et al.

The human brain possesses the extraordinary capability to contextualize the information it receives from our environment. The entorhinal-hippocampal plays a critical role in this function, as it is deeply engaged in memory processing and constructing cognitive maps using place and grid cells. Comprehending and leveraging this ability could significantly augment the field of artificial intelligence. The multi-scale successor representation serves as a good model for the functionality of place and grid cells and has already shown promise in this role. Here, we introduce a model that employs successor representations and neural networks, along with word embedding vectors, to construct a cognitive map of three separate concepts. The network adeptly learns two different scaled maps and situates new information in proximity to related pre-existing representations. The dispersion of information across the cognitive map varies according to its scale - either being heavily concentrated, resulting in the formation of the three concepts, or spread evenly throughout the map. We suggest that our model could potentially improve current AI models by providing multi-modal context information to any input, based on a similarity metric for the input and pre-existing knowledge representations.

Abner Hernandez, Tomás Arias-Vergara, Daiqi Liu et al.

Phonological features provide a language-general and linguistically grounded representation of speech. We present PhonoQ-2.0, a multilingual frame-level phonological feature recognizer built on self-supervised speech models. The system directly predicts a structured 22-dimensional feature vector per frame encoding manner, vowel quality, place, and voicing, instead of deriving features from phoneme outputs. To ensure phonologically coherent predictions, we introduce a manner-conditioned gating mechanism that activates valid feature groups. Evaluated across multiple languages and corpora, PhonoQ-2.0 achieves an average macro-F1 of 91.3% in-domain and 88.9% out-of-domain. Compared to a strong CTC phoneme baseline, it delivers consistent gains of +8.8 F1 in-domain and +8.6 out-of-domain on average. In unseen-language evaluation, PhonoQ-2.0 improves macro-F1 from 66.9% to 73.6% (+6.7 on average), with gains of up to +10.8 points.

Aline Sindel, Andreas Maier, Vincent Christlein

For art investigations of paintings, multiple imaging technologies, such as visual light photography, infrared reflectography, ultraviolet fluorescence photography, and x-radiography are often used. For a pixel-wise comparison, the multi-modal images have to be registered. We present a registration and visualization software tool, that embeds a convolutional neural network to extract cross-modal features of the crack structures in historical paintings for automatic registration. The graphical user interface processes the user's input to configure the registration parameters and to interactively adapt the image views with the registered pair and image overlays, such as by individual or synchronized zoom or movements of the views. In the evaluation, we qualitatively and quantitatively show the effectiveness of our software tool in terms of registration performance and short inference time on multi-modal paintings and its transferability by applying our method to historical prints.

Luis Carlos Rivera Monroy, Leonhard Rist, Martin Eberhardt et al.

Histopathology imaging is crucial for the diagnosis and treatment of skin diseases. For this reason, computer-assisted approaches have gained popularity and shown promising results in tasks such as segmentation and classification of skin disorders. However, collecting essential data and sufficiently high-quality annotations is a challenge. This work describes a pipeline that uses suspected melanoma samples that have been characterized using Multi-Epitope-Ligand Cartography (MELC). This cellular-level tissue characterisation is then represented as a graph and used to train a graph neural network. This imaging technology, combined with the methodology proposed in this work, achieves a classification accuracy of 87%, outperforming existing approaches by 10%.

Abner Hernandez, Paula Andrea Pérez-Toro, Juan Camilo Vásquez-Correa et al.

Collecting speech data is an important step in training speech recognition systems and other speech-based machine learning models. However, the issue of privacy protection is an increasing concern that must be addressed. The current study investigates the use of voice conversion as a method for anonymizing voices. In particular, we train several voice conversion models using self-supervised speech representations including Wav2Vec2.0, Hubert and UniSpeech. Converted voices retain a low word error rate within 1% of the original voice. Equal error rate increases from 1.52% to 46.24% on the LibriSpeech test set and from 3.75% to 45.84% on speakers from the VCTK corpus which signifies degraded performance on speaker verification. Lastly, we conduct experiments on dysarthric speech data to show that speech features relevant to articulation, prosody, phonation and phonology can be extracted from anonymized voices for discriminating between healthy and pathological speech.