EMEimo Martens

Eimo Martens

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Eimo Martens has a total of 204 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Cross-Modal Contrastive Learning of ECG and Angiography Representations for Severe Stenosis Classification

    by Nikola Cenikj, Özgün Turgut, Alexander Müller, Alexander Steger, Jan Kehrer, Marcus Brugger, Daniel Rueckert, Eimo Martens, Philip Müller

  • Echo2ECG: Enhancing ECG Representations with Cardiac Morphology from Multi-view Echos

    by Michelle Espranita Liman, Özgün Turgut, Alexander Müller, Eimo Martens, Daniel Rueckert, Philip Müller

  • Open Access

    Magnetocardiography to screen adults with arrhythmogenic cardiomyopathy: A feasibility study

    by Samuel Friese, Philipp Wunderl, Tobias Jensch, Lena Wunderl, Cordula M. Wolf, Bettina Reich, Christian Meierhofer, Reinhard Heckel, Isabel Diebold, Eimo Martens, et al.

  • Open Access

    Respiratory signal extraction from the electrocardiogram for risk stratification in cardiac patients

    by Viktoria Sanders, Ruben Alejandro Amort, Petra Barthel, Alexander Müller, Julia Allescher, Michael Dommasch, Helene Hildegard Heidegger, Eimo Martens, Manuel Rattka, Arne M Müller, et al.

    Respiration is a fundamental vital sign and an established marker of disease severity and mortality risk, particularly in patients with cardiovascular disease. Despite its prognostic significance, respiratory monitoring remains underused in routine clinical practice and remote telemonitoring. At the same time, advances in biosignal processing enable robust extraction of respiratory information from the ubiquitous electrocardiogram, allowing respiratory assessment without additional sensors. This review highlights respiration as an underused biosignal in cardiovascular medicine and summarizes physiological principles and technical approaches for electrocardiogram-derived respiration, including respiratory sinus arrhythmia, respiration-related QRS amplitude modulations, and electrical axis shifts. It emphasizes the potential of electrocardiogram-derived respiration—especially nocturnal respiratory rate—in cardiac patients to enhance risk stratification and guide clinical decision-making, including patient selection for implantable cardioverter-defibrillator therapy. The feasibility of integrating nocturnal respiratory rate monitoring into standard surveillance of cardiac patients is discussed, alongside emerging evidence that this marker may identify patients with limited benefit from prophylactic implantable cardioverter-defibrillator therapy. The technical possibility to derive respiration from routine electrocardiographic recordings represents a significant yet under-utilized opportunity. Electrocardiogram-derived respiratory rate is a promising scalable non-invasive biomarker for individualized risk stratification and may complement conventional indicators such as left-ventricular ejection fraction. Large-scale prospective randomized trials are now needed to establish whether incorporating electrocardiogram-derived respiratory markers into routine practice can improve patient outcomes and optimize therapy allocation.


  • Open Access

    Multi-View Stenosis Classification Leveraging Transformer-Based Multiple-Instance Learning Using Real-World Clinical Data

    by Nikola Cenikj, Özgün Turgut, Alexander Müller, Alexander Steger, Jan Kehrer, Marcus Brugger, Daniel Rueckert, Eimo Martens, Philip Müller

    Coronary artery stenosis is a leading cause of cardiovascular disease, diagnosed by analyzing the coronary arteries from multiple angiography views. Although numerous deep-learning models have been proposed for stenosis detection from a single angiography view, their performance heavily relies on expensive view-level annotations, which are often not readily available in hospital systems. Moreover, these models fail to capture the temporal dynamics and dependencies among multiple views, which are crucial for clinical diagnosis. To address this, we propose SegmentMIL, a transformer-based multi-view multiple-instance learning framework for patient-level stenosis classification. Trained on a real-world clinical dataset, using patient-level supervision and without any view-level annotations, SegmentMIL jointly predicts the presence of stenosis and localizes the affected anatomical region, distinguishing between the right and left coronary arteries and their respective segments. SegmentMIL obtains high performance on internal and external evaluations and outperforms both view-level models and classical MIL baselines, underscoring its potential as a clinically viable and scalable solution for coronary stenosis diagnosis. Our code is available at https://github.com/NikolaCenic/mil-stenosis.


  • Open Access

    Telemedicine use among physicians in the German outpatient sector: A secondary analysis of a cardiologist-dominated web-based survey with Bayesian model averaging and exploratory machine learning

    by Pascal Petit, Nicolas Vuillerme, Jan Gehrmann, Johannes Stephan, Felix Mühlensiepen, Jonathan Nübel, Eimo Martens, Franziska Hahn

    Background: Cardiovascular diseases remain a major health burden in Germany, and telemedicine (TM) offers promising solutions for outpatient care, yet barriers limit uptake. While prior studies relied on qualitative or conventional statistical methods, they often struggled with model uncertainty and complex relationships. Building on a national survey, this study applies Bayesian Model Averaging (BMA) and extreme gradient boosting (XGBoost). Objectives: This study aimed to explore candidate associations and patterns related to TM use among physicians in the German outpatient sector. Methods: We conducted a secondary analysis of a web-based survey carried out between 2023 and 2024. BMA was applied to identify explanatory associations with TM use, explicitly accounting for model uncertainty. XGBoost with SHAP values was used to explore classification patterns in a hypothesis-generating framework using repeated nested cross-validation. Results: Of the 165 respondents, 95 (58%) reported using TM. BMA revealed a limited number of variables with moderate to high posterior inclusion probabilities (PIP), alongside substantial overall model uncertainty, with TM use associated with receiving information from professional associations or insurers and perceived TM benefits (e.g., improving patients’ everyday quality of life, improving doctor-patient relationship). Practicing in Lower Saxony was associated with lower TM use. XGBoost demonstrated limited discriminative ability, with performance statistically indistinguishable from chance. SHAP-based analyses therefore identified exploratory patterns, including features such as information status, workplace, perceptions of TM’s benefits for patients (e.g., health literacy, compliance and adherence) and TM’s barriers (e.g., data protection, implementation incompatibility), as well as employment status. Conclusion: TM adoption in Germany’s outpatient sector appears associated with structural–economic factors and physicians’ perceptions of patient-related benefits. However, given the substantial model uncertainty in BMA and the limited predictive performance of the machine learning model, all findings should be interpreted cautiously. The machine learning component should be considered exploratory and hypothesis-generating.

  • Open Access

    Rheumatic Heart Disease in Europe: Insights from a Pilot Screening Study and a Scoping Review

    by Martina Steinmaurer, Lea Baertz, Taariq Salie, Shekhar Saha, Dominik Joskowiak, Ulrich Seybold, Eimo Martens, Joselyn Rwebembera, Christian Hagl, Mark Engel

    Background: Rheumatic heart disease (RHD) remains a preventable cause of cardiovascular morbidity/mortality, disproportionately affecting socioeconomically disadvantaged populations. While largely controlled in high-income countries, RHD persists in low- and middle-income countries and may remain underrecognized among migrants from endemic regions in Europe, where systematic surveillance data are lacking. This study aimed to explore the implementation and diagnostic yield of echocardiographic screening in a migrant population and to contextualize these findings through a scoping review of European RHD research. Methods: We conducted a single-center pilot study in Munich, Germany, screening 150 recently-arrived migrants (aged 5–26 years) from RHD-endemic regions using 2023 World Heart Federation (WHF) echocardiographic criteria. Medical history was documented. Furthermore, we conducted a scoping literature review of European studies reporting RHD published since 2000, in accordance with Cochrane and PRISMA guidelines (PROSPERO ID: CRD42024538000). Results: Among screened participants (mean age 20.0 ± 5.9 years), ten (6.8%) showed abnormalities warranting confirmatory echocardiography according to the WHF protocol for RHD diagnosis, with mitral regurgitation being the most frequent finding. No participant had a prior diagnosis of RHD. Screening was integrated into routine health examinations without any obvious disruption of routine workflows. The scoping review identified 86 publications, predominantly from Turkey and Italy, highlighting fragmented research, methodological heterogeneity, limited prevalence data, and a focus on surgical management rather than early detection. No multicenter screening studies in asymptomatic at-risk populations were identified, limiting a more accurate assessment of disease burden. Conclusions: This study suggests that systematic echocardiographic screening for RHD in high-risk migrant populations in Europe may facilitate the identification of individuals with echocardiographic abnormalities requiring confirmatory evaluation. Combined with the fragmented European literature, these findings underscore the need for harmonized surveillance, larger multicenter studies, and careful evaluation of targeted screening approaches into migrant health programs to inform future epidemiologic and implementation research. Lay Summary What question did this study address, and why does it matter? This study asked whether echocardiographic screening for rheumatic heart disease (RHD) can be implemented in migrant health assessments in Europe, and whether such screening could contribute to a better understanding of RHD in a potentially high-risk population. In addition, we sought to explore whether existing research suggests that RHD may be under-recognized in selected high-risk populations in Europe, particularly in the context of migration from endemic regions. To address this, we reviewed the current European literature to assess the extent of research activity and available evidence on RHD. Where was the study done, and what were the main results? The study was conducted in Munich, Germany, where 150 recently-arrived migrants aged 5–26 years underwent echocardiographic screening as part of a pilot program. Approximately 7% of participants showed cardiac abnormalities that may require further evaluation, although these findings do not represent confirmed diagnoses. In addition, a scoping review of 86 publications from across Europe was performed to contextualize the pilot findings within the broader European RHD literature. The literature was highly fragmented, with most studies originating from a small number of countries. Many focused on hospital-based management rather than early detection, and only a few addressed screening in at-risk populations such as migrants. What do these findings mean in practice? Given that early identification of RHD could help prevent long-term cardiac complications, echocardiographic screening may be a relevant component of routine migrant health assessments to identify individuals requiring further diagnostic evaluation. RHD may still be present in select high-risk groups in Europe, but appears to be insufficiently systematically monitored. Since current European evidence remains limited and fragmented, we recommend that larger studies be conducted to determine whether targeted screening should be implemented more widely.

  • Open Access

    Multi-View Stenosis Classification Leveraging Transformer-Based Multiple-Instance Learning Using Real-World Clinical Data

    by N. Cenikj, Ö. Turgut, A. Müller, A. Steger, J. Kehrer, M. Brugger, D. Rueckert, E. Martens, P. Müller

    Coronary artery stenosis is a leading cause of cardiovascular disease, diagnosed by analyzing the coronary arteries from multiple angiography views. Although numerous deep-learning models have been proposed for stenosis detection from a single angiography view, their performance heavily relies on expensive view-level annotations, which are often not readily available in hospital systems. Moreover, these models fail to capture the temporal dynamics and dependencies among multiple views, which are crucial for clinical diagnosis. To address this, we propose SegmentMIL, a transformer-basedmulti-viewmultiple-instance learning framework for patient-level stenosis classification. Trained on a real-world clinical dataset, using patient-level supervision and without any view-level annotations, SegmentMIL jointly predicts the presence of stenosis and localizes the affected anatomical region, distinguishing between the right and left coronary arteries and their respective segments. SegmentMIL obtains high performance on internal and external evaluations and outperforms both view-level models and classical MIL baselines, underscoring its potential as a clinically viable and scalable solution for coronary stenosis diagnosis. Our code is available at https://github.com/NikolaCenic/mil-stenosis.

  • Open Access

    Clearing the way through telemedicine: visualizing and analyzing patient flow and challenges in heart failure telemonitoring

    by Franziska Hahn, Josephine Heller, Alexander Steger, Manuel Rattka, David Federle, Jan Gehrmann, Karl-Ludwig Laugwitz, Petra Barthel, Marie Arens, Stefan Veith, et al.

    Aim: This study aims to explore the implementation steps, challenges, and optimization strategies for telemonitoring in the management of patients with heart failure, a method that holds promise for improving clinical outcomes and reducing healthcare system burdens. Subject and methods: A mixed-methods approach was employed, comprising semi-structured interviews with international physicians and an online survey conducted with German telemedicine centers. The research focused on the examination of patient flow processes and identifying barriers to effective telemonitoring, particularly in organizational structures and patient engagement. Results: The findings outline clear phases of patient flow in telemonitoring and reveal significant challenges, especially concerning patient involvement and operational management. Key areas for improvement were identified, including workflow standardization, optimization of resource allocation, and process efficiency enhancements. Conclusion: Addressing the outlined challenges and leveraging technological advancements can make telemonitoring a vital component of heart failure management. This approach could lead to reduced hospital readmissions, enhanced clinical outcomes, and significant improvements in patients’ quality of life.