25gma163 10.3205/25gma163 urn:nbn:de:0183-25gma1632 Meeting Abstract Bodmer Bodmer Nicolas S. NS

medignition AG, Zürich, Switzerland

author Lang Lang Sandra S

Universität Zürich, Medizinische Fakultät, Zürich, Switzerland

author Galushko-Jäckel Galushko-Jäckel Nina N

Universität Zürich, Medizinische Fakultät, Zürich, Switzerland

author Dupuis Dupuis Monique M

Universität Zürich, Medizinische Fakultät, Zürich, Switzerland

author Bachmann Bachmann Lucas M. LM

medignition AG, Zürich, Switzerland

author Lienhard Lienhard Kenny R. KR

medignition AG, Züirch, Switzerland

author German Medical Science GMS Publishing House

Düsseldorf

610 20250908 engl This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). M0626 163 Jahrestagung der Gesellschaft für Medizinische Ausbildung (GMA) V-22 Digitalisierung – KI und innovative Ansätze Düsseldorf 20250908 20250910 V-22-03 TextObjective: The increase in medical student enrollment in Switzerland, driven by expanded study capacities, will impose additional demands on academic supervision. To address this challenge, a digital tool has been developed at the Faculty of Medicine, University of Zurich, to equip medical students with foundational competencies in scientific writing. The tool aims to enhance students’ autonomy in academic writing while simultaneously alleviating faculty workload through adherence to standardized structural and content guidelines.Methods: In collaboration with medignition AG (Zurich, Switzerland), the Faculty of Medicine designed an integrated e-learning course complemented by an AI-driven writing assistant. The e-learning component introduces key principles of scientific writing, including manuscript structure, reporting guidelines and methodological approaches. The AI assistant provides iterative, real-time feedback on structural organization, linguistic accuracy, and content coherence during the writing process of key parts of the manuscript. Delivered as a web-based application, the system was piloted and evaluated by 24 medical students from the University of Zurich.Results: The evaluation revealed that a substantial proportion of students (75%) experienced elevated stress levels when composing their master’s thesis, with 58% identifying the thesis writing as a stressor and 42% experiencing challenges with English-language academic writing. Only 13% of participants felt adequately supported during thesis preparation. The e-learning modules were positively received, with 71% of students rating them as beneficial in preparing for thesis writing. The AI assistant proved to be beneficial, with 75% of participants finding it helpful or very helpful in streamlining the drafting process. The System Usability Scale (SUS) yielded a score of 76/100, surpassing the benchmark average of 68 . However, students expressed a desire for enhanced AI functionalities to support the development of an entire manuscript.Discussion: The e-learning platform with an AI-assisted writing tool demonstrated significant potential to support medical students in academic writing. For successful implementation, universities should proactively adopt these emerging technologies while establishing robust technical and legal frameworks. Transparent guidelines on data security, ethical considerations, and the choice between proprietary and open-source AI models are essential to ensure secure use, upholding academic standards and protecting intellectual property.Take Home Messages:AI-supported writing tools providing real-time feedback can substantially support medical students in composing their master’s thesis/dissertation.Real-time, iterative feedback from AI assistants is particularly valued for structuring content and refining grammar/flow.Universities must implement clear technical and legal standards to fully harness AI technologies in academic writing support. Brooke J 1986 System usability scale (SUS): a quick-and-dirty method of system evaluation user information 1-7 Brooke J. System usability scale (SUS): a quick-and-dirty method of system evaluation user information. Reading, UK: Digital equipment co ltd.; 1986. p.1-7. 0 0 0 0