Biomedical subjects

Dieter Hayn

Publications and source records attributed to Dieter Hayn.

3 recordsLinked to original sources

PubMed2026

Privacy-Preserving Linkage of Distributed Biological, Clinical, and Imaging Data Supporting Artificial Intelligence in Pediatric Oncology.

BACKGROUND: Cancer remains the leading cause of disease-related mortality in children over the age of one in Europe, with over 35,000 new pediatric cases and more than 6,000 deaths annually. Due to the rarity of pediatric cancers, clinical trial protocols often substitute for formal treatment guidelines, resulting in many children being enrolled in multiple trials, with biological samples and genomic data stored in various biobanks. Data collection in pediatric oncology is challenging, with sparse data acquired over extended periods, underscoring the need for optimal utilization of all available information through linked, privacy-preserving datasets. METHODS: Here, we report the development of a distributed, privacy-preserving data infrastructure for the PRIMAGE project, a European initiative aimed at supporting artificial intelligence (AI)-driven image analysis for pediatric cancer prognostics. The infrastructure leverages the European Patient Identity (EUPID) Services for Privacy-Preserving Record Linkage, enabling pseudonymized data integration across clinical, biological, and imaging sources. The system incorporates EUPID's hashing and phonetic matching protocols to pseudonymize patient identifiers and link distributed datasets, facilitating secondary data use in compliance with the General Data Protection Regulation. RESULTS: Data from over 700 neuroblastoma patients from European trials and hospitals were linked and uploaded to the PRIMAGE platform, where AI models predict clinical outcomes. CONCLUSION: This infrastructure successfully facilitated AI model development, advancing pediatric oncology research, and offering a scalable framework for future European health data initiatives, such as the European Health Data Space.

Journal Article↗

PubMed2006

Cardiac anisotropy: is it negligible regarding noninvasive activation time imaging?

The aim of this study was to quantify the effect of cardiac anisotropy in the activation-based inverse problem of electrocardiography. Differences of the patterns of simulated body surface potential maps for isotropic and anisotropic conditions were investigated with regard to activation time (AT) imaging of ventricular depolarization. AT maps were estimated by solving the nonlinear inverse ill-posed problem employing spatio-temporal regularization. Four different reference AT maps (sinus rhythm, right-ventricular and septal pacing, accessory pathway) were calculated with a bidomain theory based anisotropic finite-element heart model in combination with a cellular automaton. In this heart model a realistic fiber architecture and conduction system was implemented. Although the anisotropy has some effects on forward solutions, effects on inverse solutions are small indicating that cardiac anisotropy might be negligible for some clinical applications (e.g., imaging of focal events) of our AT imaging approach. The main characteristic events of the AT maps were estimated despite neglected electrical anisotropy in the inverse formulation. The worst correlation coefficient of the estimated AT maps was 0.810 in case of sinus rhythm. However, all characteristic events of the activation pattern were found. The results of this study confirm our clinical validation studies of noninvasive AT imaging in which cardiac anisotropy was neglected.

Action Potentials↗

PubMed2003

Development of a new QT algorithm with heterogenous ECG databases.

An algorithm for automated QT interval assessments has been developed and evaluated using the PhysioNet QT database and the electrocardiogram multilead database (2 collections of electrocardiograms with different characteristics, eg, numbers of leads and expert annotations). QRS onset and coarse T offset detection was based on the definition of a short time window, within which the range of signal amplitudes was calculated and compared to given threshold values. The final position of T offset was based on a combination of 3 methods: decreasing thresholds, multiple tangents, and a model based approach. The evaluation was based on the comparison of a waveform marker as computed automatically, and those of the human experts. Mean and standard deviation of those differences compared well to other algorithms and to inter-expert variations. Waveform marker detection was successful in at least 98% of the annotated beats in both databases, thus, indicating the robustness of the proposed method.

Algorithms↗