Expected impact 

The project is believed to increase efficiency of communication between healthcare institutions/organisations and the public. The project partners – NHS24 and Cochrane are excellent examples of institution which will benefit from this project.   

 

HimL’s impact on Public Health Services

Public services, and the people who use them, across all regions of Europe could benefit from high accuracy translation. Not only does it make these services more efficient, saving money on expensive and slow human translation, but it is easier to maintain translations of documents which are constantly changing, either due to new policies or due to constant local infrastructure changes (e.g., new doctors andclinics, etc.).

This project aims to make it easy for NHS24 to provide complete, up-to-date coverage of important sections of the NHS24 public information websites in multiple languages, greatly increasing the traffic and the impact of its content in foreign languages.

The case of best practice is being addressed by COCHRANE. The collaboration produces systematic reviews of medical research to inform health professionals, patients, policy makers and others. Cochrane reviews bring together the combined results of the world’s best medical research studies, and are widely recognised as the gold standard in evidence-based healthcare information.

Organisations such as both COCHRANE and the NHS24 provide valuable public health care information which can be useful to people across the world. The challenge that these organisations have is to maintain the highest quality, and most up-to-date information, in as many target languages as possible. If high quality translation is made available, Consumer Health Care information now presented in English will then be available to all European citizens, regardless of mother tongue and language skills.

 

Scientific and Technological Impact

The HimL project targets translation from English into four languages (Romanian, German, Czech and Polish) which have a range of different resources, but have relatively low translation quality. We bring together advances in domain adaptation, treatment of morphology and semantic fidelity in order to create translation systems which can be deployed for public health information. In close connection with other researchers in the MT community, we will attempt to steer the field and re-focus on adequacy of MT. HimL contributes to and puts to test methods addressing top MT challenges: domain adaptation, target-side fluency, and core semantic fidelity.