The European Cystic Fibrosis Society (ECFS) produces peer-reviewed e-learning modules intended to support clinical teams and centers. These resources are predominantly available in English, which limits access for clinicians and centers in many countries. The source identifies cost and time as primary barriers to producing high-quality manual translations of educational content. The study set out to pilot an alternative approach using AI to generate translations and to evaluate the feasibility and perceived quality of AI-assisted subtitles for ECFS modules.
An AI translation program was used to create subtitles for ECFS peer-reviewed education modules. The source does not report the specific AI program or software used. For each target language, two independent native-language speakers with extensive cystic fibrosis healthcare experience were identified to review the AI output. Their tasks were to review, edit, and validate the subtitles produced by the AI system. Following editing and validation, an online evaluation survey was developed and circulated to collect user views on the quality of the final translated modules.
The project produced education packages composed of six subtitled modules for each of three languages: Ukrainian, Romanian, and Turkish. The source emphasizes that corrections by the independent native-language reviewers were essential for each language package. The exact nature and number of corrections, the time required for editorial review, and the specific workflow steps are not detailed in the source material.
Evaluation of the subtitled modules took place in two countries. An online evaluation survey, designed to assess user perceptions of translation quality and utility, was circulated after the review process. The source reports that eighteen completed surveys were received. The content of the survey instrument, response rate relative to the number of recipients, respondent characteristics, and precise survey metrics or scoring scales were not reported in the source.
According to the source, aggregated evaluation results indicated high levels of accuracy for the final translated modules after expert review. Feedback from respondents was described as very positive regarding the utility of the subtitled education packages. The study therefore reports that AI-generated translations, when followed by expert native-language review and correction, produced usable educational modules in the three pilot languages.
The article reports several important outcomes but omits specific operational details that would be relevant for replication or deeper assessment. Not reported in the source were:
These gaps limit the ability to judge generalizability, cost savings, time efficiencies, or reproducibility beyond the pilot described.
The pilot indicates that AI can be used to produce initial translations of ECFS e-learning modules and that subsequent expert native-language review is essential to reach a high-quality final product. Final modules in Ukrainian, Romanian, and Turkish were perceived by respondents as accurate and useful. The approach shows promise as a partial solution to the documented cost and time barriers to translating educational materials for non-English speaking Cystic Fibrosis centers. However, important implementation details and quantitative evaluation data were not provided in the source, so further work with transparent reporting on the AI tool, editorial effort, cost, and objective quality metrics will be needed to confirm scalability and impact.
Additional source details
This work is reported in Pediatr Pulmonol (2026 Aug;61(8):e71764) with PMID 42530427, PMCID PMC13422125, and DOI 10.1002/ppul.71764. The authors and affiliations span multiple European institutions and patient organizations. The study focused on pilot translation and evaluation; full reproducibility information was not reported in the abstract.