Marie-Christine Pali

Dr. Marie-Christine Pali

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 April 2022 in IEEE Signal Processing Letters
Verified
Journal Article
Closed access
MPMarie-Christine Pali
Marie-Christine Pali
SRSimon Ruetz
Simon Ruetz
KSKarin Schnass
Karin Schnass
This paper studies the performance of OMP in comparison to Thresholding in the case in which only a perturbed version of the generating dictionary is known. Both theory and numerical simulations show that simple Thresholding is a viable alternative to OMP in applications where only a perturbed version of the generating dictionary is given, such as dictionary learning.

4 April 2023 in Information and Inference: A Journal of the IMA
Verified
Journal Article
Closed access
PDF file
MPMarie-Christine Pali
Marie-Christine Pali
KSKarin Schnass
Karin Schnass
This paper studies the convergence behaviour of dictionary learning via the Iterative Thresholding and K-residual Means (ITKrM) algorithm. On one hand, it is proved that ITKrM is a contraction under much more relaxed conditions than previously necessary. On the other hand, it is shown that there seem to exist stable fixed points that do not correspond to the generating dictionary, which can be characterised as very coherent. Based on an analysis of the residuals using these bad dictionaries, replacing coherent atoms with carefully designed replacement candidates is proposed. In experiments on synthetic data, this outperforms random or no replacement and always leads to full dictionary recovery. Finally, the question how to learn dictionaries without knowledge of the correct dictionary size and sparsity level is addressed. Decoupling the replacement strategy of coherent or unused atoms into pruning and adding, and slowly carefully increasing the sparsity level, leads to an adaptive version of ITKrM. In several experiments, this adaptive dictionary learning algorithm is shown to recover a generating dictionary from randomly initialized dictionaries of various sizes on synthetic data and to learn meaningful dictionaries on image data.

10 July 2025 in arXiv
Verified
Preprint
Open access
PDF file
MPMarie-Christine Pali
Marie-Christine Pali
CSChristina Schwaiger
Christina Schwaiger
MGMalik Galijasevic
Malik Galijasevic
VLValentin K. Ladenhauf
Valentin K. Ladenhauf
SMStephanie Mangesius
Stephanie Mangesius
EGElke Ruth Gizewsk
Elke Ruth Gizewsk
The analysis of carotid arteries, particularly plaques, in multi-sequence Magnetic Resonance Imaging (MRI) data is crucial for assessing the risk of atherosclerosis and ischemic stroke. In order to evaluate metrics and radiomic features, quantifying the state of atherosclerosis, accurate segmentation is important. However, the complex morphology of plaques and the scarcity of labeled data poses significant challenges. In this work, we address these problems and propose a semi-supervised deep learning-based approach designed to effectively integrate multi-sequence MRI data for the segmentation of carotid artery vessel wall and plaque. The proposed algorithm consists of two networks: a coarse localization model identifies the region of interest guided by some prior knowledge on the position and number of carotid arteries, followed by a fine segmentation model for precise delineation of vessel walls and plaques. To effectively integrate complementary information across different MRI sequences, we investigate different fusion strategies and introduce a multi-level multi-sequence version of U-Net architecture. To address the challenges of limited labeled data and the complexity of carotid artery MRI, we propose a semi-supervised approach that enforces consistency under various input transformations. Our approach is evaluated on 52 patients with arteriosclerosis, each with five MRI sequences. Comprehensive experiments demonstrate the effectiveness of our approach and emphasize the role of fusion point selection in U-Net-based architectures. To validate the accuracy of our results, we also include an expert-based assessment of model performance. Our findings highlight the potential of fusion strategies and semi-supervised learning for improving carotid artery segmentation in data-limited MRI applications.

19 September 2026 in European Journal of Neurology
Verified
Journal Article
Open access
PDF file
KBKlaus Berek
Klaus Berek
MSMartin Schmidauer
Martin Schmidauer
FFFabian Föttinger
Fabian Föttinger
MAMichael Auer
Michael Auer
TBThomas Berger
Thomas Berger
HHHarald Hegen
Harald Hegen
Background: The relationship between intrathecal kappa free light chain (κ‐FLC) synthesis and retinal layer atrophy in multiple sclerosis (MS) is unknown. Objective: To investigate whether the κ‐FLC index is associated with peripapillary retinal nerve fiber layer (pRNFL) and ganglion cell plus inner plexiform layer (GCIPL) thickness. Methods: Patients with newly diagnosed clinically isolated syndrome or relapsing MS and available cerebrospinal fluid (CSF) analysis and optical coherence tomography (OCT) were included. Clinical and magnetic resonance imaging (MRI) data were also assessed. κ‐FLC concentrations were measured by nephelometry, and the κ‐FLC index was calculated as (CSF κ‐FLC/serum κ‐FLC)/albumin quotient. Results: A total of 100 patients at a median age of 33 (25th–75th percentile: 25–39) years and a female predominance (58%) were included. In multivariable linear regression analysis adjusted for age, sex, disease duration, number of T2‐hyperintense and number of contrast‐enhancing MRI lesions, the κ‐FLC index was associated with pRNFL (log‐transformed κ‐FLC index: β : −2.51; 95% confidence interval [CI]: −4.01, −1.01; p = 0.001) and GCIPL thinning (log‐transformed κ‐FLC index: β : −1.24, 95% CI: −2.32, −0.17; p = 0.024). Conclusions: Intrathecal plasma cell activity, as measured by the κ‐FLC index, is associated with neuroaxonal damage, as reflected by pRNFL and GCIPL atrophy.

9 September 2026 in Diagnostics
Verified
Journal Article
Open access
PDF file
MPMarie-Christine Pali
Marie-Christine Pali
SMStephanie Mangesius
Stephanie Mangesius
CEConstantin Emanuel Eisenschink
Constantin Emanuel Eisenschink
PDPhilipp Deisl
Philipp Deisl
LNLukas Neumann
Lukas Neumann
AGAstrid E. Grams
Astrid E. Grams
Objectives: We aimed to determine quantitative photon-counting CT (PCCT)-derived iodine map (IM) attenuation associated with visual detectability of parenchymal hyperattenuation on conventional brain window (BW) images after mechanical thrombectomy (MT) and to assess corresponding relative BW attenuation changes. Methods: In this retrospective single-center study, 12 patients with anterior circulation large vessel occlusion underwent MT followed by post-interventional PCCT. BW, IM, and virtual non-contrast reconstructions were generated. Follow-up non-contrast CT served as reference for final infarction extent. Of 54 ASPECTS regions with final infarction, 40 showed IM hyperattenuation and were included in the detectability analysis. Two blinded readers assessed visual detectability. IM attenuation and relative BW attenuation increase (%ΔHU) were evaluated using ROC analysis with clustered bootstrap resampling. Thresholds were determined using the Youden index. Results: Across all 54 regions with final infarction, median IM attenuation was significantly higher in infarcted than contralateral regions (6.39 vs. 2.18 HU, p < 0.001), while absolute BW attenuation did not differ significantly. Among the 40 regions with IM hyperattenuation, 15 were visually detectable on BW images. ROC analysis showed excellent performance for IM (AUC = 0.997) and %ΔHU (AUC = 0.984). The optimal IM threshold in the original cohort was 8.33 HU (sensitivity 100%, specificity 96.0%), and the optimal %ΔHU threshold was 3.06% (sensitivity 100%, specificity 92.0%). Bootstrap-derived median thresholds were 11.69 HU (95% interval 8.33–12.67) for IM and 3.06% (3.06–20.16%) for %ΔHU. Inter-reader agreement was excellent (κ = 0.81–0.86). Conclusions: PCCT-derived iodine quantification enables objective assessment of visual detectability of post-interventional parenchymal hyperattenuation. In this cohort, an internally derived IM cutoff of 8.33 HU was associated with BW visibility. These preliminary findings require validation in larger, independent multicenter cohorts.

19 September 2026 in Diagnostics
Verified
Journal Article
Open access
PDF file
PDPhilipp Deisl
Philipp Deisl
SMStephanie Mangesius
Stephanie Mangesius
MPMarie-Christine Pali
Marie-Christine Pali
MLMaximilian Lutz
Maximilian Lutz
MKMichael Prof. Knoflach
Michael Prof. Knoflach
EGElke Ruth Gizewsk
Elke Ruth Gizewsk
Background: Intracranial arterial calcification (IAC), commonly detected on CT/CTA during ischemic stroke work-up, may reflect atherosclerotic burden relevant to recurrence. We evaluated whether CTA-detected IAC was associated with subsequent ischemic cerebral events. Methods: This retrospective single-center cohort included patients with ischemic stroke or transient ischemic attack admitted to a tertiary stroke unit between 2010 and 2023. IAC was assessed on index-event CTA using volumetric quantification and the Babiarz score. Time to first recurrent ischemic cerebrovascular event was analyzed using multivariable Cox regression. Robustness was evaluated through adjustment for stroke etiology and CTA acquisition parameters, penalized regression, bootstrap internal validation, and competing-risk analysis accounting for death. Results: Among 556 patients (median age, 71.5 years [IQR, 62.3–78.6]; 354 male), 60 (10.8%) experienced recurrence during a median follow-up of 3.5 years (IQR, 1.8–7.5). Median IAC volume was higher in patients with recurrence than in those without recurrence (45.0 mm3 [IQR, 14.1–143.2] vs. 18.3 mm3 [IQR, 0–105.6]; p = 0.01). A one-unit increase in log10(IAC + 1), corresponding to a tenfold increase in IAC + 1, was associated with a higher recurrence hazard (HR, 1.63; 95% CI, 1.18–2.24; p = 0.003). The model had a C-index of 0.727, and the association remained consistent across sensitivity analyses. Babiarz scores correlated strongly with volumetric IAC measurements in the intracranial carotid and vertebral arteries (ρ ≥ 0.93; p < 0.01). Conclusions: Greater CTA-derived IAC burden was associated with a higher hazard of recurrent ischemic cerebrovascular events. Quantitative IAC assessment warrants further investigation as a potential imaging marker of recurrence risk, although external validation is required before clinical implementation.