Searcharxiv⌕ Search

arXiv subjects

Robert Hoepner

Publications and source records attributed to Robert Hoepner.

3 recordsLinked to original sources

SWIMSS: An Open Multi-Center SwinUNETR Model for Multiple Sclerosis White Matter Lesion Segmentation

Purpose: To develop and validate SWIMSS (SWInUNETR for Multiple Sclerosis white matter lesion Segmentation), a SwinUNETR-based deep learning (DL) model for reliable automated white matter (WM) lesion segmentation, in a large multicentric cohort of people with multiple sclerosis (pwMS). Methods: A real-world longitudinal dataset of 1,385 pwMS from seven Swiss clinical centers was used to train, validate, and test SWIMSS and two state-of-the-art DL architectures: UNETR and MSSegDiff. Inference results obtained with LST-AI were also compared. Preprocessed magnetization-prepared rapid acquisition gradient echo and fluid-attenuated inversion recovery MRI sequences, acquired with various MR systems and field strengths, served as model inputs. Model performance was evaluated using the Dice similarity coefficient (DSC) with 95% confidence intervals (CI). Results: SWIMSS achieved the highest performance, with a DSC of 81.37 (95% CI: 80.63-82.11), outperforming UNETR (79.06; 95% CI: 78.33-79.79), MSSegDiff (75.70; 95% CI: 74.97-76.43), and LST-AI (63.19; 95% CI: 62.08-64.30). Normalized DSC followed a similar trend, with SWIMSS reaching 84.88. Lesion detection F1 achieved 81.14. Statistical analyses confirmed its superior performance (p < 0.001). Conclusion: SWIMSS achieved state-of-the-art performance for WM lesion segmentation. By making this model publicly available, we enable broader access to high-quality segmentation tools, allowing smaller centers, often lacking large annotated datasets or computational resources, to benefit from a model trained on a diverse, large-scale cohort. This promotes more equitable access to advanced imaging analysis and supports reproducibility and standardization in MS research.

eess.IV↗

Speech-Based Depressive Mood Detection in the Presence of Multiple Sclerosis: A Cross-Corpus and Cross-Lingual Study

Depression commonly co-occurs with neurodegenerative disorders like Multiple Sclerosis (MS), yet the potential of speech-based Artificial Intelligence for detecting depression in such contexts remains unexplored. This study examines the transferability of speech-based depression detection methods to people with MS (pwMS) through cross-corpus and cross-lingual analysis using English data from the general population and German data from pwMS. Our approach implements supervised machine learning models using: 1) conventional speech and language features commonly used in the field, 2) emotional dimensions derived from a Speech Emotion Recognition (SER) model, and 3) exploratory speech feature analysis. Despite limited data, our models detect depressive mood in pwMS with moderate generalisability, achieving a 66% Unweighted Average Recall (UAR) on a binary task. Feature selection further improved performance, boosting UAR to 74%. Our findings also highlight the relevant role emotional changes have as an indicator of depressive mood in both the general population and within PwMS. This study provides an initial exploration into generalising speech-based depression detection, even in the presence of co-occurring conditions, such as neurodegenerative diseases.

cs.CL↗

Learning from spatially inhomogenous data: resolution-adaptive convolutions for multiple sclerosis lesion segmentation

In the setting of clinical imaging, differences in between vendors, hospitals and sequences can yield highly inhomogeneous imaging data. In MRI in particular, voxel dimension, slice spacing and acquisition plane can vary substantially. For clinical applications, therefore, algorithms must be trained to handle data with various voxel resolutions. The usual strategy to deal with heterogeneity of resolution is harmonization: resampling imaging data to a common (usually isovoxel) resolution. This can lead to loss of fidelity arising from interpolation artifacts out-of-plane and downsampling in-plane. We present in this paper a network architecture designed to be able to learn directly from spatially heterogeneous data, without resampling: a segmentation network based on the e3nn framework that leverages a spherical harmonic, rather than voxel-grid, parameterization of convolutional kernels, with a fixed physical radius. Networks based on these kernels can be resampled to their input voxel dimensions. We trained and tested our network on a publicly available dataset assembled from three centres, and on an in-house dataset of Multiple Sclerosis cases with a high degree of spatial inhomogeneity. We compared our approach to a standard U-Net with two strategies for handling inhomogeneous data: training directly on the data without resampling, and resampling to a common resolution of 1mm isovoxels. We show that our network is able to learn from various combinations of voxel sizes and outperforms classical U-Nets on 2D testing cases and most 3D testing cases. This shows an ability to generalize well when tested on image resolutions not seen during training. Our code can be found at: http://github.com/SCAN-NRAD/e3nn\_U-Net.

eess.IV↗