arXiv subjects

Alexander Koch

Publications and source records attributed to Alexander Koch.

9 recordsLinked to original sources

arXiv2026

HemExp: Clinically-Guided Latent Diffusion for Modeling Hematoma Expansion

Hematoma expansion (HE) after spontaneous intracerebral hemorrhage (ICH) is a major determinant of acute triage and treatment decisions in neurosurgical care. However, most existing methods provide either a binary expansion risk or a single follow-up volume, limiting uncertainty-aware decisions. We introduce HemExp, a clinically-guided latent diffusion model that generates patient-specific follow-up non-contrast CT images, along with segmentations of intraparenchymal and intraventricular hemorrhage. Generation is conditioned on baseline imaging, clinical variables, and an explicit expansion indicator, enabling controllable simulation of realistic clinical scenarios. HemExp uses a hemorrhage-aware multi-head variational autoencoder and models progression as the difference between baseline and follow-up latent representations with a conditional diffusion model. The model is trained on paired scans from 450 patients across multiple centers and evaluated on 107 patients from a held-out institution. HemExp produces spatial HE probability maps by generating multiple synthetic follow-up images per patient to estimate distributions of plausible follow-up hematoma volumes. Perturbing clinical inputs such as symptom-onset-to-imaging time or anticoagulant status shifts the predicted follow-up volume distribution. HemExp extends binary predictors and demonstrates robust estimation of clinically relevant outcomes in the imaging space, such as hematoma volume, intraventricular involvement, and mass effects. Overall, our results support controllable latent diffusion as a promising direction for uncertainty-aware modeling of early ICH progression.

cs.CV↗

arXiv2025

RELICT: A Replica Detection Framework for Medical Image Generation

Despite the potential of synthetic medical data for augmenting and improving the generalizability of deep learning models, memorization in generative models can lead to unintended leakage of sensitive patient information and limit model utility. Thus, the use of memorizing generative models in the medical domain can jeopardize patient privacy. We propose a framework for identifying replicas, i.e. nearly identical copies of the training data, in synthetic medical image datasets. Our REpLIca deteCTion (RELICT) framework for medical image generative models evaluates image similarity using three complementary approaches: (1) voxel-level analysis, (2) feature-level analysis by a pretrained medical foundation model, and (3) segmentation-level analysis. Two clinically relevant 3D generative modelling use cases were investigated: non-contrast head CT with intracerebral hemorrhage (N=774) and time-of-flight MR angiography of the Circle of Willis (N=1,782). Expert visual scoring was used as the reference standard to assess the presence of replicas. We report the balanced accuracy at the optimal threshold to assess replica classification performance. The reference visual rating identified 45 of 50 and 5 of 50 generated images as replicas for the NCCT and TOF-MRA use cases, respectively. Image-level and feature-level measures perfectly classified replicas with a balanced accuracy of 1 when an optimal threshold was selected for the NCCT use case. A perfect classification of replicas for the TOF-MRA case was not possible at any threshold, with the segmentation-level analysis achieving a balanced accuracy of 0.79. Replica detection is a crucial but neglected validation step for the development of generative models in medical imaging. The proposed RELICT framework provides a standardized, easy-to-use tool for replica detection and aims to facilitate responsible and ethical medical image synthesis.

eess.IV↗

arXiv2024

Cross-modality image synthesis from TOF-MRA to CTA using diffusion-based models

Cerebrovascular disease often requires multiple imaging modalities for accurate diagnosis, treatment, and monitoring. Computed Tomography Angiography (CTA) and Time-of-Flight Magnetic Resonance Angiography (TOF-MRA) are two common non-invasive angiography techniques, each with distinct strengths in accessibility, safety, and diagnostic accuracy. While CTA is more widely used in acute stroke due to its faster acquisition times and higher diagnostic accuracy, TOF-MRA is preferred for its safety, as it avoids radiation exposure and contrast agent-related health risks. Despite the predominant role of CTA in clinical workflows, there is a scarcity of open-source CTA data, limiting the research and development of AI models for tasks such as large vessel occlusion detection and aneurysm segmentation. This study explores diffusion-based image-to-image translation models to generate synthetic CTA images from TOF-MRA input. We demonstrate the modality conversion from TOF-MRA to CTA and show that diffusion models outperform a traditional U-Net-based approach. Our work compares different state-of-the-art diffusion architectures and samplers, offering recommendations for optimal model performance in this cross-modality translation task.

eess.IV↗

arXiv2023

The TopCoW Challenge -- Topology-Aware Circle of Willis Segmentation for CT and MR Angiography

The Circle of Willis (CoW) is an important network of arteries connecting major circulations of the brain. Its vascular architecture is believed to influence the risk, severity, and outcome of serious neurovascular diseases. However, characterizing the highly variable CoW anatomy remains a manual and time-consuming expert task. The CoW is commonly imaged by two non-invasive angiographic imaging modalities, magnetic resonance angiography (MRA) and computed tomography angiography (CTA), yet few datasets with annotated CoW anatomy exist, and there have been no established benchmarks for comparing CoW segmentation algorithms. We organized the TopCoW benchmark challenge alongside the release of an annotated CoW dataset with 125 paired MRA and CTA scans from the same patients. Voxel-level annotations for 13 vessel components were created using virtual reality technology and verified by clinical experts. Participants submitted algorithms for CoW segmentation and variant classification, which we evaluated on internal and external test sets comprising 226 scans from over five centers. The benchmark includes voxel-level segmentation, CoW component detection, CoW variant classification, and two clinical application tasks. We received submissions from over 250 participants across six continents. Top-performing teams achieved over 90% Dice scores for CoW segmentation, over 80% F1 scores for detecting key vessel components, and over 70% balanced accuracy in CoW variant classification across nearly all test sets. The best algorithms also supported clinically relevant downstream tasks by accurately classifying fetal-type posterior cerebral arteries and localizing aneurysms in relation to CoW anatomy. This benchmark demonstrated the utility of CoW segmentation algorithms for some downstream clinical applications with explainability.

cs.CV↗

arXiv2022

Heavily doped zinc oxide with plasma frequencies in the telecommunication wavelength range

We demonstrate heavy and hyper doping of ZnO by a combination of gallium (Ga) ion implantation using a focused ion beam (FIB) system and post-implantation laser annealing. Ion implantation allows for the incorporation of impurities with nearly arbitrary concentrations, and the laser-annealing process enables dopant activation close to or beyond the solid-solubility limit of Ga in ZnO. We achieved heavily doped ZnO:Ga with free-carrier concentrations of ~10^21 cm^(-3), resulting in a plasma wavelength of 1.02 um, which is substantially shorter than the telecommunication wavelength of 1.55 um. Thus, our approach enables the control of the plasma frequency of ZnO from the far infrared down to 1.02 um, providing a promising plasmonic material for applications in this regime.

physics.app-ph↗

arXiv2022

Meta-Reinforcement Learning via Language Instructions

Although deep reinforcement learning has recently been very successful at learning complex behaviors, it requires a tremendous amount of data to learn a task. One of the fundamental reasons causing this limitation lies in the nature of the trial-and-error learning paradigm of reinforcement learning, where the agent communicates with the environment and progresses in the learning only relying on the reward signal. This is implicit and rather insufficient to learn a task well. On the contrary, humans are usually taught new skills via natural language instructions. Utilizing language instructions for robotic motion control to improve the adaptability is a recently emerged topic and challenging. In this paper, we present a meta-RL algorithm that addresses the challenge of learning skills with language instructions in multiple manipulation tasks. On the one hand, our algorithm utilizes the language instructions to shape its interpretation of the task, on the other hand, it still learns to solve task in a trial-and-error process. We evaluate our algorithm on the robotic manipulation benchmark (Meta-World) and it significantly outperforms state-of-the-art methods in terms of training and testing task success rates. Codes are available at \url{https://tumi6robot.wixsite.com/million}.

cs.RO↗

arXiv2022

Tuning carrier density and phase transitions in oxide semiconductors using focused ion beams

We demonstrate spatial modification of the optical properties of thin-film metal oxides, zinc oxide and vanadium dioxide as representatives, using a commercial focused ion beam (FIB) system. Using a Ga+ FIB and thermal annealing, we demonstrated variable doping of a band semiconductor, zinc oxide (ZnO), achieving carrier concentrations from 10^18 cm-3 to 10^20 cm-3. Using the same FIB without subsequent thermal annealing, we defect-engineered a correlated semiconductor, vanadium dioxide (VO2), locally modifying its insulator-to-metal transition (IMT) temperature by range of ~25 degrees C. Such area-selective modification of metal oxides by direct writing using a FIB provides a simple, mask-less route to the fabrication of optical structures, especially when multiple or continuous levels of doping or defect density are required.

cond-mat.mtrl-sci↗

arXiv2019

On orbit performance of the GRACE Follow-On Laser Ranging Interferometer

The Laser Ranging Interferometer (LRI) instrument on the Gravity Recovery and Climate Experiment (GRACE) Follow-On mission has provided the first laser interferometric range measurements between remote spacecraft, separated by approximately 220 km. Autonomous controls that lock the laser frequency to a cavity reference and establish the 5 degree of freedom two-way laser link between remote spacecraft succeeded on the first attempt. Active beam pointing based on differential wavefront sensing compensates spacecraft attitude fluctuations. The LRI has operated continuously without breaks in phase tracking for more than 50 days, and has shown biased range measurements similar to the primary ranging instrument based on microwaves, but with much less noise at a level of $1\,{\rm nm}/\sqrt{\rm Hz}$ at Fourier frequencies above 100 mHz.

astro-ph.IM↗

arXiv2013

Graphs with Plane Outside-Obstacle Representations

An \emph{obstacle representation} of a graph consists of a set of polygonal obstacles and a distinct point for each vertex such that two points see each other if and only if the corresponding vertices are adjacent. Obstacle representations are a recent generalization of classical polygon--vertex visibility graphs, for which the characterization and recognition problems are long-standing open questions. In this paper, we study \emph{plane outside-obstacle representations}, where all obstacles lie in the unbounded face of the representation and no two visibility segments cross. We give a combinatorial characterization of the biconnected graphs that admit such a representation. Based on this characterization, we present a simple linear-time recognition algorithm for these graphs. As a side result, we show that the plane vertex--polygon visibility graphs are exactly the maximal outerplanar graphs and that every chordal outerplanar graph has an outside-obstacle representation.

cs.CG↗