JEJulius Emons

Julius Emons

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Julius Emons has a total of 158 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Vulva und Vagina

    by Matthias W. Beckmann, Julius Emons, Paul Gaß, Sven Becker, Diethelm Wallwiener

  • Open Access

    The Potential of Large Language Models in Gynecologic Oncology Telephone Triage

    by Iason Psilopatis, Julius Emons, Matteo Pavone, Tibor A. Zwimpfer

    Introduction: Nurse-led telephone triage, frequently guided by the United Kingdom Oncology Nursing Society (UKONS) tool, is a cornerstone of supportive oncology care, yet the appropriateness of triage decisions varies across settings. Large Language Models (LLMs) show promise in medical communication but remain unvalidated for autonomous clinical decision-making, particularly in gynecologic oncology, where remote and telephone-based care pose distinct challenges. Methods and Materials: Twenty fictitious case vignettes representing common treatment-related toxicities in gynecologic oncology were developed. Three board-certified gynecologic consultants independently assigned UKONS Red–Amber–Green (RAG) triage categories, reaching unanimous consensus that served as the reference standard. Vignettes were submitted verbatim, as single-turn queries with no follow-up prompting, to three publicly available LLMs (OpenEvidence, ChatGPT, Google Gemini). LLM responses were mapped to RAG categories and compared descriptively against the reference standard, stratified by treatment type and platform. Results: All three LLMs achieved 100% concordance with the reference standard for all Red cases across both treatment groups, with no high-acuity case downgraded to self-care. Discrepancies clustered at lower-acuity boundaries. In the chemotherapy-alone group, errors were limited to omission of an explicit 24-h follow-up window for 1 Amber-rated complaints. In the combined group, all LLMs over-triaged intermediate-risk complaints toward immediate or same-day evaluation; low-risk complaints showed more variable over-triage. Conclusions: LLMs identified unambiguous emergent gynecologic oncology symptoms, preserving the UKONS safety floor, but showed relatively reduced precision and a conservative over-triage bias at lower-acuity margins, especially in the combined-therapy collective. These findings support a supervised, adjunctive role for LLMs, rather than autonomous deployment, pending prospective validation.

  • Open Access

    Beyond a single ratio: an adaptable deep learning approach to analyze the prognostic value of TSP distribution in ovarian cancer

    by Daniel Firmbach, Petr Kuritcyn, Michaela Benz, Pamela L Strissel, Reiner Strick, Matthias W Beckmann, Laura Glossner, Julius Emons, Volker Bruns, Arndt Hartmann, et al.

    The tumor–stroma percentage (TSP) has been proposed as a prognostic factor in ovarian cancer. When assessed manually by pathologists, the TSP can be subject to significant inter‐rater variability. Automated methods for TSP determination have recently been developed. Existing approaches typically yield a single TSP value for a relatively small region of interest in the slide, introducing bias through region selection and potentially overlooking the TSP distribution across the entire histologically examined area. We present an adaptable deep learning pipeline, efficiently fine‐tuned for ovarian cancer tissue analysis using a few‐shot learning approach, requiring only minimal new training annotations. Instead of calculating a single TSP value, our method quantifies the global TSP distribution across the entire tumor and therefore captures its spatial heterogeneity. To identify a prognostic biomarker, we systematically evaluated a panel of metrics describing this distribution in 98 patients. Our analysis revealed that the mean–median difference of the TSP distribution [hazard ratio (HR) = 0.60, 95% CI = 0.46–0.79, p < 0.001], a quantitative measure of asymmetry, together with the mean TSP (HR = 0.65, 95% CI = 0.43–0.97, p = 0.03), are biomarkers significantly associated with recurrence‐free survival. By successfully adapting a pipeline for ovarian cancer, we demonstrate that automated analysis of the entire TSP distribution provides significant prognostic information superior to that obtained by manual assessment. Spatial heterogeneity of the TSP is a key prognostic factor. © 2026 Fraunhofer IIS and The Author(s). The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

  • Open Access

    The Role of Large Language Models in Hormonal Contraception Consultation

    by Iason Psilopatis, Julius Emons, Tibor A. Zwimpfer

    Background: Large Language Models (LLMs) demonstrate promise in medical applications, but their performance in hormonal contraception consultation remains underexplored. Objective: To evaluate the accuracy and comprehensiveness of LLM-generated hormonal contraception counseling compared to evidence-based guidelines. Methods: Ten fictitious clinical case scenarios representing common contraceptive counseling situations were presented to Chat-GPT, Google Gemini, and OpenEvidence. Cases assessed medical eligibility screening, contraindication recognition, drug interactions, side effect management, and emergency contraception guidance. Two board-certified obstetrician–gynecologists independently evaluated the responses based on international clinical guidelines. Results: Across the ten predefined clinical scenarios, all three LLMs achieved high accuracy in identifying contraindications according to Medical Eligibility Criteria (MEC) and provided appropriate alternative contraceptive recommendations. Models demonstrated high proficiency in managing drug interactions, particularly the lamotrigine–estrogen interaction, and provided evidence-based side effect management strategies. However, the communication styles differed. Chat-GPT emphasized structured consultation and shared decision-making, Gemini provided practical action-oriented guidance, and OpenEvidence delivered concise evidence-focused summaries. Conclusions: In this limited set of standardized fictitious cases, the evaluated LLMs generally provided responses that were broadly aligned with selected guideline recommendations. Current LLMs require cautious deployment, given limitations in individualized assessment and health literacy optimization, and are best positioned as complementary educational tools rather than replacements for professional contraceptive counseling.

  • Open Access

    The Role of Large Language Models in the Promotion of Minimally Invasive Interventional Radiologic Methods in Gynecology and Obstetrics

    by Iason Psilopatis, Julius Emons, Kleio Vrettou, Tibor A. Zwimpfer

    Background: Minimally invasive interventional radiology (IR) offers effective, uterus-preserving treatments for several gynecologic and obstetric conditions such as uterine fibroids, adenomyosis and postpartum hemorrhage. Despite their efficacy, these methods remain underused, partly to limited awareness among clinicians and patients. Large language models (LLMs) may help bridge this gap by providing accessible, reliable information. Objective: To evaluate how current LLMs address knowledge gaps and promote awareness of minimally invasive IR methods in gynecology and obstetrics. Methods: A structured ten-question instrument was used to query three publicly available LLMs (OpenEvidence, ChatGPT, and Google Gemini). Responses were analyzed for accuracy, completeness, safety considerations, and patient-centered communication. Results: All three models accurately identified a range of medical, minimally invasive, and surgical treatments for uterine fibroids, adenomyosis, and postpartum hemorrhage, with OpenEvidence and ChatGPT providing more detailed and clinically nuanced responses. OpenEvidence achieved the highest scores overall, closely followed by ChatGPT, while Google Gemini scored lower, particularly in completeness and patient-centered communication. In more complex scenarios, performance differences became more pronounced, with OpenEvidence again leading, ChatGPT performing strongly, and Google Gemini lagging behind. Overall, OpenEvidence and ChatGPT demonstrated higher accuracy, completeness, and safety considerations, whereas Google Gemini showed comparatively weaker and less consistent performance. Conclusions: LLMs may endorse the promotion of minimally invasive IR methods in gynecology and obstetrics, but their outputs vary considerably in quality. Ongoing refinement and integration of evidence-based sources are essential before routine use in clinical practice. Therefore, effective collaboration between artificial intelligence (AI) developers and medical professionals is essential to harness this technology’s full potential.

  • Open Access

    Leveraging artificial intelligence for evidence‐based recommendations in uterine fibroid therapy: Addressing the unmet need in German healthcare—A clinical trial

    by Iason Psilopatis, Laura Lotz, Nadezda Sipulina, Felix Heindl, Georgia Levidou, Julius Emons

    Objective: To evaluate the potential of an artificial intelligence (AI)‐driven large language model, ChatGPT 4.0, to provide personalized, evidence‐based treatment recommendations for uterine fibroids. Methods: ChatGPT 4.0 was trained using evidence‐based data from Uptodate and German medical literature. The algorithm generated individualized recommendations based on clinical characteristics and patient preferences. Usability and quality were assessed through questionnaires completed by 40 gynecologists and 45 women with fibroids. Results: Most gynecologists found the algorithm user‐friendly and comprehensive, with 15 expressing a willingness to integrate it into practice and 24 acknowledging its potential to enhance healthcare efficiency. Although only half believed it would improve patient outcomes, the tool was generally well received. Patients found the algorithm easy to understand and helpful for exploring treatment options, with the majority feeling it empowered informed discussions with their healthcare providers. A minority expressed dissatisfaction with usability or helpfulness. Conclusion: ChatGPT 4.0 offers a promising AI‐driven tool for personalized fibroid management in the absence of formal guidelines. Although not a substitute for official recommendations, it could support clinical decision making and enhance patient education. Further integration with standardized guidelines and prospective trials is needed to optimize its clinical utility.


  • Open Access

    Update Gynecologic Malignancies 2025 – Expert Opinion on Systemic Therapy for Early and Advanced Gynecological Cancers

    by Julius Emons, Julia Gocke, Manuel Hörner, Patrik Pöschke, Carla Schulmeyer, Frederik Alexander Stübs, Annika Krückel, Niklas Amann, Matthias W. Beckmann

    There have been major changes in the understanding of gynecologic malignancies in recent years, leading to new therapy options and subsequently to greater responsibilities for every professional treating those patients. The most significant therapeutic advances were achieved with checkpoint inhibitors (CPI), especially for endometrial and cervical cancer. In ovarian cancer the dominant and most important new substances are poly (ADP-ribose) polymerase inhibitors (PARPi). This review aims to summarize the latest studies and developments in the therapeutic landscape of endometrial, ovarian, and cervical cancer. The treatment of advanced endometrial cancer has changed significantly with the introduction of CPI such as dostarlimab (RUBY trial), durvalumab (DUO-E trial) and pembrolizumab (Keynote-868 trial). For ovarian cancer PARPi have shown substantial PFS benefits in key approval trials, including PRIMA for niraparib, PAOLA for olaparib, and ATHENA-MONO for rucaparib. These findings have established PARPi as the standard of care in maintenance therapy. Overall survival (OS) data for PRIMA and PAOLA are now available and are analyzed and placed into context in this article. Furthermore, mirvetuximab soravtansine is the first antibody-drug conjugate (ADC) approved in Germany for platinum-resistant ovarian cancer for patients with folate receptor alpha expression. The Keynote-A18 and BEATcc trials have opened new options for the utilization of immuno-oncology in cervical cancer treatment. Along with new therapeutic options, new biomarkers have also become part of daily clinical practice as predictive and prognostic factors as well as forming the basis for targeted personalized medicine.The use of CPI is revolutionizing the treatment of all gynecologic cancers and offers significant benefits for progression-free survival (PFS) and OS in most therapy regimens. With the increased use of ADCs, this is not the end of these developments. Therapy algorithms from a certified German oncology center are developed and presented in this article.


  • Open Access

    Cross-sectional study on user requirements for developing a digital patient navigator app

    by Katharina Seitz, Manuela Langbein, Chloë Goossens, Stefanie Altmannshofer, Peter A. Fasching, Carolin C. Hack, Felix Heindl, Matthias W. Beckmann, Julius Emons, Hanna Huebner

    Objective: This study aimed to assess patients’ preferences regarding the content and features of a digital patient navigator app. A secondary objective was to explore how patient characteristics influence the perceived importance and anticipated frequency of use of different app components. Methods: The study was a monocentric cross-sectional survey conducted at the University Hospital Erlangen. A questionnaire consisting of 20 questions was designed to inquire about patients’ preferences for a digital patient navigator. Participants had breast diseases. Descriptive analyses were conducted to identify key app features and their correlation with patient characteristics. Results: Questionnaires from 243 patients were analyzed. Preferred key app features, in order of preference from highest to lowest, were as follows: a communication feature to chat with treating physicians, a personalized treatment plan, real-time notifications, mediation services for providing support, 3D navigation within the hospital, and personal documents. Patients’ age, native language, and education significantly influenced the assessment of importance and frequency of use of some app features. A personalized treatment plan was considered more important by younger individuals (≤60 years, p < 0.001), whereas real-time notifications were considered more important by participants with a lower educational level ( p = 0.003) and younger individuals ( p = 0.036). Increased frequency of use of a personalized treatment plan tool was also associated with younger age ( p < 0.001) and lower education levels ( p = 0.025). Conclusion: These findings suggest that a patient navigator app could be a valuable tool for a broad range of patients, potentially complementing in-person patient navigators. To ensure broad usability and acceptance, future development should account for varying needs across patient subgroups—particularly in terms of personalization, language accessibility, and communication features.


  • Personal Profiling im Bereich der Senologie

    by Annika S. Behrens, Matthias W. Beckmann, Peter A. Fasching, Hanna Huebner, Julius Emons

    Zusammenfassung Hintergrund Der Begriff der personalisierten Medizin wird immer relevanter. Die Möglichkeiten der Diagnostik umfassen hierbei nicht nur genetische und molekulare Tumorprofile, sondern auch den Einsatz präziser und individueller bildgebender Methoden. Fragestellung Die Entwicklung und Implementation geeigneter diagnostischer Verfahren mit hoher Sensitivität und Spezifität, welche gleichzeitig auf die individuellen Risikofaktoren und biologischen Merkmale der Patientin abgestimmt sind, bleibt eine Herausforderung. Material und Methoden Um ein Personal Profiling zu ermöglichen, muss eine umfassende Diagnostik etabliert werden, welche sämtliche Parameter wie Bildgebung, molekulare und genetische Marker sowie Real-World-Daten und den Einsatz künstlicher Intelligenz (KI) berücksichtigt. Dieser Artikel beschäftigt sich mit verschiedenen Ansätzen der personalisierten Diagnostik beim Mammakarzinom und beleuchtet den aktuellen klinischen Standard, innovative Forschungsbereiche und die sich daraus ergebenden Herausforderungen. Diskussion Hürden neuerer bildgebender Verfahren sind vor allem die Standardisierung der Bildanalyse sowie die Validierung dieser Verfahren in großen klinischen Studien. Der Einsatz von KI erfordert nicht nur entsprechendes technisches und medizinisches Know-how, sondern auch einen sensiblen Umgang mit Themen wie Datenschutz und Privatsphäre der Patientinnen. Real-World-Register bieten Einblicke in die reale Behandlungssituation und sind daher von großer Bedeutung.

  • Open Access

    The role of artificial intelligence in gynecologic and obstetric emergencies

    by Iason Psilopatis, Felix Heindl, Susanne Cupisti, Ulrike Fischer, Valentina Kohlmann, Michael Schneider, Simon Bader, Annika Krueckel, Julius Emons