Searcharxiv⌕ Search

arXiv subjects

Sven Bergmann

Publications and source records attributed to Sven Bergmann.

15 recordsLinked to original sources
arXiv2026

Robust retinal biometrics for patient identity verification and retrieval across age and imaging devices

Patient identity errors can compromise longitudinal medical records, research databases, and downstream clinical decisions. We present a retinal biometric system for verifying claimed identities and retrieving the correct identity from color fundus images. We trained a 512-dimensional metric-learning encoder combining a ConvNeXtV2 backbone with ArcFace and triplet losses on 227,004 images from 21,851 patient-eye identities in the Rotterdam Study, spanning multiple imaging devices and up to 32.6 years of follow-up. The system was evaluated on held-out Rotterdam Study data and externally on the UK Biobank and Age-Related Eye Disease Study (AREDS). Before evaluation, we used the model to screen for identity inconsistencies and manually adjudicated flagged images, identifying incorrect assignments in 0.588% of Rotterdam Study images, 0.259% of UK Biobank images, and 0.164% of AREDS images. In retrospective-only verification after removing near-duplicate images, the system achieved AUROCs of 0.9998, 0.9997, and 0.9998 in the Rotterdam Study, UK Biobank, and AREDS, respectively. For identity retrieval using only previously acquired images, Recall@1 was 99.7%, 97.2%, and 97.6%, respectively, from galleries averaging 4436-8510 identities; the correct identity appeared among the top five results in at least 98.6% of cases. Performance remained robust across imaging devices and long follow-up intervals, while lower image quality and inconsistent retinal fields accounted for most failures. These findings establish retinal anatomy as a durable biometric signal, useful for safeguarding the integrity of longitudinal imaging records.

cs.CV↗
arXiv2026

retinalysis-vascx: An explainable software toolbox for the extraction of retinal vascular biomarkers

Automatic extraction of retinal vascular biomarkers from color fundus images (CFI) is crucial for large-scale studies of the retinal vasculature. We present VascX, an open-source Python toolbox that extracts biomarkers from CFI artery-vein segmentations. VascX starts from vessel segmentation masks, extracts their skeletons, builds undirected and directed vessel graphs, and resolves vessel segments into longer vessels. A comprehensive set of biomarkers is derived, including vascular density, central retinal equivalents (CREs), and tortuosity. Spatially localized biomarkers may be calculated over grids placed relative to the fovea and optic disc. VascX is released via GitHub and PyPI with comprehensive documentation and examples. Our test-retest reproducibility analysis on repeat imaging of the same eye by different devices shows that most VascX biomarkers have moderate to excellent agreement (ICC > 0.5), with important differences in the level of robustness of different biomarkers. Our analyses of biomarker sensitivity to image perturbations and heuristic parameter values support these differences and further characterize VascX biomarkers. Ultimately, VascX provides an explainable and easily modifiable feature-extraction toolbox that complements segmentation to produce reliable retinal vascular biomarkers. Our graph-based biomarker computation stages support reproducible, region-aware measurements suited for large-scale clinical and epidemiological research. By enabling easy extraction of existing biomarkers and rapid experimentation with new ones, VascX supports oculomics research. Its robustness and computational efficiency facilitate scalable deployment in large databases, while open-source distribution lowers barriers to adoption for ophthalmic researchers and clinicians.

q-bio.TO↗
arXiv2013

The hourglass and the early conservation models - co-existing evolutionary patterns in vertebrate development

Developmental constraints have been postulated to limit the space of feasible phenotypes and thus shape animal evolution. These constraints have been suggested to be the strongest during either early or mid-embryogenesis, which corresponds to the early conservation model or the hourglass model, respectively. Conflicting results have been reported, but in recent studies of animal transcriptomes the hourglass model has been favored. Studies usually report descriptive statistics calculated for all genes over all developmental time points. This introduces dependencies between the sets of compared genes, and may lead to biased results. Here we overcome this problem using an alternative modular analysis. We used the Iterative Signature Algorithm to identify distinct modules of genes co-expressed specifically in consecutive stages of zebrafish development. We then performed a detailed comparison of several gene properties between modules, allowing for a less biased and more powerful analysis. Notably, our analysis corroborated the hourglass pattern only at the regulatory level, with sequences of regulatory regions being most conserved for genes expressed in mid-development, but not at the level of gene sequence, age or expression, in contrast to some previous studies. The early conservation model was supported with gene duplication and birth that were the most rare for genes expressed in early development. Finally, for all gene properties we observed the least conservation for genes expressed in late development or adult, consistent with both models. Overall, with the modular approach, we showed that different levels of molecular evolution follow different patterns of developmental constraints. Thus both models are valid, but with respect to different genomic features.

q-bio.PE↗
arXiv2002

The Iterative Signature Algorithm for the analysis of large scale gene expression data

We present a new approach for the analysis of genome-wide expression data. Our method is designed to overcome the limitations of traditional techniques, when applied to large-scale data. Rather than alloting each gene to a single cluster, we assign both genes and conditions to context-dependent and potentially overlapping transcription modules. We provide a rigorous definition of a transcription module as the object to be retrieved from the expression data. An efficient algorithm, that searches for the modules encoded in the data by iteratively refining sets of genes and conditions until they match this definition, is established. Each iteration involves a linear map, induced by the normalized expression matrix, followed by the application of a threshold function. We argue that our method is in fact a generalization of Singular Value Decomposition, which corresponds to the special case where no threshold is applied. We show analytically that for noisy expression data our approach leads to better classification due to the implementation of the threshold. This result is confirmed by numerical analyses based on in-silico expression data. We discuss briefly results obtained by applying our algorithm to expression data from the yeast S. cerevisiae.

physics.bio-ph↗
arXiv2001

Constraining Models of New Physics in Light of Recent Experimental Results on $a_{ψK_S}$

We study extensions of the Standard Model where the charged current weak interactions are governed by the CKM matrix and where all tree-level decays are dominated by their Standard Model contribution. We constrain both analytically and numerically the ratio and the phase difference between the New Physics and the Standard Model contributions to the mixing amplitude of the neutral $B$ system using the experimental results on $R_u$, $Δm_{d,s}$, $ε_K$ and $a_{ψK_S}$. We present new results concerning models with minimal flavor violation and update the relevant parameter space. We also study the left-right symmetric model with spontaneously broken CP, probing the viability of this model in view of the recent results for $a_{ψK_S}$ and other observables.

hep-ph↗
arXiv2001

Generic Smooth Connection Functions - A New Analytic Approach to Interpolation

We present a generic solution to the fundamental problem of how to connect two points in a plane by a smooth curve that goes through these points with a given slope. The smoothness of any curve depends both on its curvature and its length. The smoothest curves correspond to a particular compromise between minimal curvature and minimal length. They can be described by a class of functions that satisfy certain boundary conditions and minimize a weight functional. The value of this functional is given essentially by the average of the curvature raised to some power νtimes the length of the curve. The parameter νdetermines the importance of minimal curvature with respect to minimal length. In order to find the functions that obtain the minimal weight, we use extensively notions that are well-known in classical mechanics. The minimization of the weight functional via the Euler-Lagrange formalism leads to a highly non-trivial differential equation. Using the symmetries of the problem it is possible to find conserved quantities, that help to simplify the problem to a level where the solution functions can be written in a closed form for any given ν. Applying the appropriate coordinate transformation to these solutions allows to adjust them to all possible boundary conditions.

physics.class-ph↗
arXiv2000

Standard Model correlations between K -> πν\barνdecays and observables in B Physics

Including the recent preliminary results of BaBar and BELLE experiments, we update the currently allowed intervals for various CKM parameters: $\barρ$, $\barη$, $\sin2β$, $\sin2α$, $\sin^2γ$. We also update the SM prediction for the rates of the $K_L \to π^0 ν\barν$ and $K^+ \to π^+ ν\barν$ decays, their ratio $a_{πν\barν}$, as well as for certain observables related to $B$ Physics like the CP asymmetries $a_{ψK_S}$ in $B^0_d \to J/ψK_S$ and $a_{ψϕ}$ in $B^0_s \to J/ψϕ$ or the mass differences $Δm_q$ ($q=d,s$) in the $B^0_q-\bar B^0_q$ systems. We investigate the correlations between them. The strongest correlations are between i) $a_{πν\barν}$ and $a_{ψK_S}$, ii) $\BR(K^+ \to π^+ ν\barν)$ and $Δm_d/Δm_s$ and iii) $\BR(K_L \to π^0 ν\barν)$ and $a_{ψϕ}$. These correlations are likely to be violated in the presence of New Physics and therefore provide stringent tests of the Standard Model.

hep-ph↗
arXiv2000

Lessons from CLEO and FOCUS Measurements of D0-anti-D0 Mixing Parameters

If the true values of the D0-anti-D0 mixing parameters lie within the one sigma ranges of recent measurements, then there is strong evidence for a large width difference, y > 0.01, and large SU(3) breaking effects in strong phases, δ> pi/4. These constraints are model independent, and would become stronger if M_{12}/Gamma_{12} << 1 in the D0-anti-D0 system. The interesting fact that the FOCUS result cannot be explained by a large mass difference is not trivial and depends on the small D0-anti-D0 production asymmetry in FOCUS and the bounds on CP violating effects from CLEO. The large value of δmight help explain why y ~ sin^2(theta_c).

hep-ph↗
arXiv2000

Lepton number violating interactions and their effects on neutrino oscillation experiments

Mixing between bosons that transform differently under the standard model gauge group, but identically under its unbroken subgroup, can induce interactions that violate the total lepton number. We discuss four-fermion operators that mediate lepton number violating neutrino interactions both in a model-independent framework and within supersymmetry (SUSY) without R-parity. The effective couplings of such operators are constrained by: i) the upper bounds on the relevant elementary couplings between the bosons and the fermions, ii) by the limit on universality violation in pion decays, iii) by the data on neutrinoless double beta decay and, iv) by loop-induced neutrino masses. We find that the present bounds imply that lepton number violating neutrino interactions are not relevant for the solar and atmospheric neutrino problems. Within SUSY without R-parity also the LSND anomaly cannot be explained by such interactions, but one cannot rule out an effect model-independently. Possible consequences for future terrestrial neutrino oscillation experiments and for neutrinos from a supernova are discussed.

hep-ph↗
arXiv1999

Neutrino propagation in matter with general interactions

We present a general analysis of the effective potential for neutrino propagation in matter, assuming a generic set of Lorentz invariant non-derivative interactions. We find that in addition to the known vector and axial vector terms, in a polarized medium also tensor interactions can play an important role. We compute the effective potential arising from a tensor interaction. We show that the components of the tensor potential transverse to the direction of the neutrino propagation can induce a neutrino spin-flip, similar to the one induced by a transverse magnetic field.

hep-ph↗
arXiv1999

New Physics Effects in Doubly Cabibbo Suppressed D Decays

The most sensitive experimental searches for $D-\bar D$ mixing use $D^0 \to K^+ π^-$ decays. It is often assumed that effects of New Physics and, in particular, CP violation, can appear through the mixing, while the $c \to d u \bar s$ decay amplitude cannot have significant contributions from New Physics and is, therefore, CP conserving to a good approximation. We examine this assumption in two ways. First, we calculate the contributions to the decay in various relevant models of New Physics: Supersymmetry without R-parity, multi-scalar models, left-right symmetric models, and models with extra quarks. We find that phenomenological constraints imply that the New Physics contributions are indeed small compared to the standard model doubly Cabibbo suppressed amplitude. Second, we show that many of our constraints hold model-independently. We find, however, one case where the model-independent bound is rather weak and a CP violating contribution of order 30% is not excluded.

hep-ph↗
arXiv1999

Can lepton flavor violating interactions explain the atmospheric neutrino problem?

We investigate whether flavor changing neutrino interactions (FCNIs) can be sufficiently large to provide a viable solution to the atmospheric neutrino problem. Effective operators induced by heavy boson exchange that allow for flavor changing neutrino scattering off quarks or electrons are related by an $SU(2)_L$ rotation to operators that induce anomalous tau decays. Since $SU(2)_L$ violation is small for New Physics at or above the weak scale, one can use the upper bounds on lepton flavor violating tau decays or on lepton universality violation to put severe, model-independent bounds on the relevant non-standard neutrino interactions. Also $Z$-induced flavor changing neutral currents, due to heavy singlet neutrinos, are too small to be relevant for the atmospheric neutrino anomaly. We conclude that the FCNI solution to the atmospheric neutrino problem is ruled out.

hep-ph↗
arXiv1998

Can lepton flavor violating interactions explain the LSND results?

If the atmospheric and the solar neutrino problem are both explained by neutrino oscillations, and if there are only three light neutrinos, then all mass-squared differences between the neutrinos are known. In such a case, existing terrestrial neutrino oscillation experiments cannot be significantly affected by neutrino oscillations, but, in principle there could be an anomaly in the neutrino flux due to new neutrino interactions. We discuss how a non-standard muon decay $μ^+ \to e^+ \barν_e ν_\ell$ would modify the neutrino production processes of these experiments. Since $SU(2)_L$ violation is small for New Physics above the weak scale one can use related flavor-violating charged lepton processes to constrain these decays in a model independent way. We show that the upper bounds on $μ\to 3e$, muonium-antimuonium conversion and $τ\to μe e$ rule out any observable effect for the present experiments due to $μ^+ \to e^+ \barν_e ν_\ell$ for $\ell=e,μ,τ$, respectively. Applying similar arguments to flavor-changing semi-leptonic reactions we exclude the possibility that the "oscillation signals" observed at LSND are due to flavor-changing interactions that conserve total lepton number.

hep-ph↗
arXiv1998

Z-induced FCNCs and their effects on Neutrino Oscillations

Adding singlet neutrinos to the standard model spectrum in general gives rise to Z-induced flavor-changing neutral currents. We study the impact of these currents on matter-induced neutrino oscillations in the sun and in supernovae. While the effects for solar neutrinos are negligible, dramatic effects are possible for supernova neutrinos.

hep-ph↗