ANAnatol-Fiete Näher

Anatol-Fiete Näher

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Anatol-Fiete Näher has a total of 55 co-authors. Here are the top co-authors by number of shared publications:

  • Felix Balzer

    Institute of Medical Informatics, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Berlin, Germany

    6 shared publications
  • Marvin Kopka

    Division of Ergonomics, Department of Psychology and Ergonomics (IPA), Technische Universität Berlin, Berlin, Germany

    5 shared publications
  • Matthias Schulte-Althoff

    Institute of Medical Informatics, Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Berlin, Germany

    5 shared publications
  • Andreas Edel

    Department of Anesthesiology and Intensive Care Medicine, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Berlin, Germany

    3 shared publications
  • Elias Grünewald

    Institute of Medical Informatics, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Berlin, Germany

    3 shared publications
  • Falk Meyer-Eschenbach

    Institute of Medical Informatics, Charité – Universitätsmedizin Berlin, Germany

    3 shared publications

Latest Publications

  • Open Access

    Effect of Data Preprocessing Decisions on Sepsis-3 Labels in MIMIC-III and eICU-CRD: A Sensitivity Analysis

    by Falk Meyer-Eschenbach, Johannes V. Stoephasius, Anatol-Fiete Näher, Fabian Kohlbacher, Severin Kohler, Elias Grünewald, Nicolas Frey, Andreas Edel, Jon Barrenetxea, Felix Balzer, et al.

    Objective: Published Sepsis-3 detection rates range from 3.4 % to 65.2 % in MIMIC-III and from 9.8 % to 47.9 % in eICU-CRD. We tested whether data preprocessing decisions alone produce differences of this magnitude when data, cohort and sepsis definition are held fixed. Materials and Methods: Using the MIMIC-III and eICU-CRD databases we developed and applied a configurable pipeline (medalabel) that stepwise maps the databases into the OMOP Common Data Model, standardizes and cleans the data, and preprocesses them to compute Sepsis-3 labels. Each setting of the preprocessing represents one decision of published Sepsis-3 implementations, and settings can be freely combined. On fixed cohorts of first ICU stays (13,388 in MIMIC-III, 94,782 in eICU-CRD), we changed one setting of a reference configuration at a time (51 variants in five preprocessing domains) and compared detection rates and septic stays (Jaccard index) with the reference. We approximately reconstructed six published pipelines and combined the settings with the largest effects in pairs to test whether their effects add up. The time a value is measured can differ from the time it is documented and thus available for automated analysis. Beyond the usual retrospective view based on measurement time, the pipeline optionally combines both time points to simulate a real-time system. Results: The reference configuration labelled 31.5 % of ICU stays in MIMIC-III and 8.2 % in eICU-CRD as septic. Changing a single setting resulted in sepsis detection rates of 15.3-58.6 % in MIMIC-III and 4.6-17.2 % in eICU-CRD. The largest shifts from the reference, all in the same direction in both databases, came from the SOFA baseline, up to +27.1 percentage points (pp) in MIMIC-III and +9.0 pp in eICU-CRD, from disabling imputation (-16.3 and -2.5 pp), the temporal window (+12.6 and +5.3 pp) and aggregation (-10.1 and -0.8 pp); clinician-validated parameter sets had little effect (-4.8 and -1.1 pp). Even variants that changed the rate by less than 2 pp labelled partly different patients, with Jaccard indices as low as 0.67 in MIMIC-III and 0.56 in eICU-CRD (1 = identical septic stays). Reconstructions deviated from the published rates by 2-10 pp. Two settings applied together deviated from the sum of their single effects by 0.9-11.8 pp in MIMIC-III and by up to 4.9 pp in eICU-CRD. Simulating a real-time system lowered the MIMIC-III rate by 5.0 pp and delayed the label in 45 % of the stays that remained septic. Discussion: With the infection criterion held fixed, preprocessing decisions alone can produce a considerable part of the variation between published sepsis detection rates. Similar rates can stem from different patients, which makes studies and prediction models difficult to compare, and retrospective labels rely on values that a real-time system would not yet have. Conclusion: Sepsis-3 labels from health records are only comparable if their data preprocessing is known. Configurable pipelines allow quantifying the effect of these decisions, computing alternative labels on the same data and, in future, validating them against clinician-adjudicated cases.

  • Open Access

    Variability in automated sepsis case detection: a systematic analysis of implementation methods in clinical data repositories

    by Falk Meyer-Eschenbach, René Schmiedler, Johannes von Stoephasius, Camille Zhang, Luis Kronfli, Nicolas Frey, Anatol-Fiete Näher, Jacques Ehret, Johanna Nothacker, Christof von Kalle, et al.

    Objective: To systematically review and characterize methodological heterogeneity in sepsis case detection using the Medical Information Mart for Intensive Care III (MIMIC-III) and eICU-CRD databases. Materials and Methods: We conducted a PRISMA-guided systematic review of PubMed and Web of Science (publication years 2016-2024). Studies were eligible if they applied the SOFA-based Sepsis-3 definition to MIMIC-III or eICU-CRD, provided at least one explicit label-calculation detail, reported a traceable detection rate, and met further criteria detailed in the PRISMA flow diagram. We classified each study by study type and unit of analysis (patients, admissions, or ICU stays) and extracted methodological details on sepsis case detection across 6 domains. For studies with available source code, we additionally examined code structure and repository dependencies to identify methodological decisions across these domains. Results: Of 396 publications screened, 64 met inclusion criteria and 12 provided source code. Sepsis detection rates ranged from 3.4% to 65.2% in MIMIC-III and from 9.8% to 47.9% in eICU-CRD. Substantial variability persisted among studies using identical cohort definitions within both databases (MIMIC-III: 16.9%-42.2%; eICU-CRD: 13.9%-31.4%). In a generalized linear mixed model, between-study heterogeneity was extreme (overall τ²≈0.79; 95% prediction interval for a new study’s detection rate 5%–63%), and study-level characteristics—including database, study type, infection detection method, code availability, and unit of analysis—explained only a small proportion of this variance (≈9%), with no characteristic reaching statistical significance. The overall proportion of studies reporting methodological details varied by domain: SOFA calculation (53.1%), infection detection methods (42.2%), temporal windows (37.5%), aggregation methods (26.6%), and missing-data handling (17.2%). Source code analysis identified 321 implementation decisions, revealing heterogeneity in baseline SOFA definitions (SOFA = 0 vs dynamic baseline), temporal windows (infection-centered vs ICU-admission-centered) and infection detection methods (antibiotic-culture matching vs APACHE-based diagnosis). Dependencies among several MIMIC-III repositories suggested propagation of implementation decisions across studies. Discussion: Clinically validated sepsis definitions yield substantially different detection rates across studies using identical datasets. Statistical modeling showed that this variability is driven almost entirely by between-study implementation differences rather than sampling error or study characteristics. Conclusion: To improve reproducibility in sepsis research and the robustness of sepsis prediction models, we recommend standardized reporting of sepsis case detection methodology and the publication of version-controlled source code.

  • Open Access

    Review of: "Advancing Personal Health Data Spaces: Governance, Interoperability, and Real-World Evidence for Learning Healthcare Systems"

    by Anatol Fiete Näher

    This commentary provides a well-thought-out conceptual framework for Personal Health Data Spaces (PHDS) and integrates governance, interoperability, AI, and real-world evidence into a patient-centered model. Its strongest contribution is the clear di...

  • Open Access

    Tossing a Coin: Preferences for Rank-Ordering in Triage Decisions

    by Peer Keßler, Anatol-Fiete Näher, Ivar Krumpal

    The discussion about the fair allocation of scarce medical resources through triage decisions gained momentum during the COVID-19 pandemic. Ethical experts proposed to maximize the overall treatment benefits by prioritizing patients with the highest estimated post-treatment length of life, and apply a random lottery if the estimated post-treatment length of life for two patients is equal. However, implementing expert ethics requires the procedures to at least coincide with the preferences of the general public to ensure trust in institutions and social cohesion. Although there is a growing body of research on preferences of the public in triage decisions, evidence for preferences in accordance with this rank-ordering approach is lacking. To address this research gap, 1,998 English-speaking adults from North America and Europe completed an online conjoint experiment. We find a close match between the rank-ordering approach and participants’ preferences. We discuss implications and provide guidance for future research.

  • Open Access

    Addressing Algorithmic Bias Through Data Disclosure: Evidence from a Randomized Experiment Among Physicians

    by Louis Agha-Mir-Salim, Anna Lena Fehlhaber, Nicolas Frey, Ivar Krumpal, Anatol-Fiete Näher

    Machine learning (ML) based clinical decision support systems (CDSS) are increasingly being developed for acute care, yet clinicians must judge whether algorithmic recommendations are applicable to their patient populations. Bias in medical ML is often attributed to unrepresentative validation data, but empirical evidence is lacking on how transparency about dataset demographics influences clinical adoption decisions — an issue central to emerging AI regulation. We conducted a randomized vignette experiment among 257 physicians evaluating ML-based CDSS for acute coronary syndrome diagnosis, systematically varying the age and gender composition of validation datasets. Physicians rated perceived clinical relevance and indicated whether they would use each system in practice. Demographic mismatches between validation data and physicians’ patient populations substantially reduced intended CDSS use. A 10-percentage-point underrepresentation of patients aged over 65 years decreased intended use by 15 percentage points (95% confidence interval (CI): -16.8 to -13.8, P<0.001), while overrepresentation of female patients showed analogous effects. Responses were asymmetric and dimension-specific: physicians penalized underrepresentation of older patients but not overrepresentation, and overrepresentation of female patients but not underrepresentation, indicating uneven sensitivity to demographic bias across dimensions. Effects were consistent across physician characteristics and increased with mismatch magnitude, suggesting systematic assessment of model applicability. These findings provide first experimental evidence that demographic transparency shapes physician adoption of ML-based CDSS and highlight the need to align transparency requirements with how clinicians interpret demographic disclosures in clinical decision-making.


  • Open Access

    Variability in Automated Sepsis Case Detection: A Systematic Analysis of Implementation Methods in Clinical Data Repositories

    by Falk Meyer-Eschenbach, René Schmiedler, Johannes V. Stoephasius, Camille Zhang, Luis Kronfli, Nicolas Frey, Anatol-Fiete Näher, Jacques Ehret, Johanna Nothacker, Christof von Kalle, et al.

    Objective: To systematically review and characterize methodological heterogeneity in sepsis case detection using the MIMIC-III and eICU-CRD databases. Materials and Methods: We conducted a PRISMA-guided systematic review of PubMed and Web of Science (publication years 2016–2024). We extracted methodological details on sepsis case detection across six domains: parameter coverage, temporal windows, aggregation methods, missing-data handling, SOFA calculation, and infection detection methods. For studies with available source code, we additionally examined code structure and repository dependencies to identify methodological decisions across these domains. Results: Of 396 publications screened, 64 met the inclusion criteria and 12 provided available source code. Sepsis detection rates ranged from 3.4% to 65.2% in MIMIC-III and from 9.8% to 47.9% in eICU-CRD. Substantial variability persisted among studies using identical cohort definitions within both databases (MIMIC-III: 16.9%–42.2%; eICU-CRD: 13.9%–31.4%). The overall proportion of studies reporting methodological details varied by domain: SOFA calculation (53.1%), infection detection methods (42.2%), temporal windows (37.5%), aggregation methods (26.6%) and missing-data handling (17.2%). Source code analysis identified 321 implementation decisions, revealing heterogeneity in baseline SOFA definitions (SOFA=0 vs dynamic baseline), temporal windows (infection-centered vs ICU-admission-centered) and infection detection methods (antibiotic-culture matching vs APACHE-based diagnosis). Dependencies among several MIMIC-III repositories suggested propagation of implementation decisions across studies. Discussion: Clinically validated sepsis definitions yield substantially different detection rates across studies using identical datasets, indicating heterogeneity in computational implementation. Conclusion: To improve reproducibility in sepsis research and the robustness of sepsis prediction models, we recommend standardized reporting of sepsis case detection methodology and the publication of version-controlled source code.


  • Open Access

    More Digital, More Diverse? How Digital Health Technologies Contribute to Representativeness of Clinical Trials

    by Anatol-Fiete Näher, Linea Schmidt, Marvin Kopka, Matthias Schulte-Althoff, Akhyar Ahmed

    The rapid adoption of digital health technologies (DHT) in clinical trials promises to enhance trial representativeness by eliminating geographical barriers and reducing participant burden. However, their actual impact on participant representation remains poorly understood. Our analysis of 68,206 US-based trials from ClinicalTrials.gov uncovers gaps in participant representation in trials utilizing DHT for data collection (DHTcd). While DHTcd trials achieved overrepresentation of female participants (+1.7% versus US Census data and +5.5% vs. non-DHT trials) and reduced underrepresentation of Black participants by 4.2% (vs. non-DHT trials), they showed underrepresentation of adults aged 65 years and older (−8.8% vs. non-DHT trials). These effects vary by therapeutic area. We furthermore find that DHTcd trials report group-specific sample compositions less frequently than non-DHT trials, suggesting that growing DHTcd adoption has not been systematically used to improve trial representativeness. We conclude that evidence-based and indication-specific implementation strategies are needed to enhance standards for DHT-based data collection in clinical trials and to prevent a widening digital divide in trial participation.

  • Open Access

    On the Fidelity versus Privacy and Utility Trade-Off of Synthetic Patient Data

    by Tim Adams, Colin Birkenbihl, Karen Otte, Hwei Geok Ng, Jonas Adrian Rieling, Anatol-Fiete Naeher, Ulrich Sax, Fabian Prasser, Holger Froehlich, Alzheimer’s Disease Neuroimaging Initiative

    Summary: The use of synthetic data is a widely discussed and promising solution for privacy-preserving medical research. Synthetic data may however not always be privacy preserving and can vary greatly in terms of data fidelity and utility. We systematically evaluate the trade-offs between privacy, fidelity, and utility across five synthetic data models and three patient-level datasets. We evaluate fidelity based on statistical similarity to the real data, utility on three machine learning use cases and privacy via membership inference, singling out, and attribute inference risks. Synthetic data without differential privacy (DP) maintained fidelity and utility without evident privacy breaches, whereas DP-enforced models significantly disrupted correlation structures. K-anonymity-based data sanitization, while preserving fidelity, introduced notable privacy risks. Our findings emphasize the need to advance methods that effectively balance privacy, fidelity, and utility in synthetic patient data generation. Highlights: Differential Privacy (DP) had a detrimental effect on feature correlations Models that did not implement DP showed good fidelity compared to real data Non-DP synthetic models showed no strong evidence of privacy breaches k-anonymization produced high fidelity data but showed notable privacy risks


  • Open Access

    Generalizability in real‐world trials

    by Anatol‐Fiete Näher, Marvin Kopka, Felix Balzer, Matthias Schulte‐Althoff

    Real‐world evidence (RWE) trials have a key advantage over conventional randomized controlled trials (RCTs) due to their potentially better generalizability. High generalizability of study results facilitates new biological insights and enables targeted therapeutic strategies. Random sampling of RWE trial participants is regarded as the gold standard for generalizability. Additionally, the use of sample correction procedures can increase the generalizability of trial results, even when using nonrandomly sampled real‐world data (RWD). This study presents descriptive evidence on the extent to which the design of currently planned or already conducted RWE trials takes sampling into account. It also examines whether random sampling or procedures for correcting nonrandom samples are considered. Based on text mining of publicly available metadata provided during registrations of RWE trials on clinicaltrials.gov, EU‐PAS, and the OSF‐RWE registry, it is shown that the share of RWE trial registrations with information on sampling increased from 65.27% in 2002 to 97.43% in 2022, with a corresponding increase from 14.79% to 28.30% for trials with random samples. For RWE trials with nonrandom samples, there is an increase from 0.00% to 0.95% of trials in which sample correction procedures are used. We conclude that the potential benefits of RWD in terms of generalizing trial results are not yet being fully realized.


  • Open Access

    Measuring fairness preferences is important for artificial intelligence in health care

    by Anatol-Fiete Näher, Ivar Krumpal, Esther-Maria Antão, Erika Ong, Marina Rojo, Fred Kaggwa, Felix Balzer, Leo Anthony Celi, Katarina Braune, Lothar H Wieler, et al.