DIA SESSION RECAP: Automated Translations for Pharma Companies
Last week at the DIA Europe Annual Meeting in Vienna, TransPerfect—together with ICON Clinical Research—led an “Automated Translations Workshop” to discuss how companies can transform the translation process using technology and artificial intelligence to work more efficiently and get drugs to market faster. As the industry becomes more global, companies are dealing with larger and larger sets of data and documents that are required in a wider variety of languages. As the volume of content grows, it becomes even more important for companies to evaluate their translation processes in order to produce more global content, while utilizing fewer resources. By introducing efficiency into the translation process, pharma companies can work to more quickly enroll patients in trials and bring new drugs to market.
The translation process for clinical trials, regulatory affairs, and other high-risk pharmaceutical documentation is historically manual and time-consuming. There are three key areas where technology can be introduced in order to bring significant time and resource reduction to the process today:
-
Using Component Content Management Systems (CCMS) to author more structured and consistent English content.
This reduces authoring time for medical writers, creates more consistent source materials, and ultimately reduces the downstream costs of the materials being translated.
-
Implementing translation workflow technology to automate the exchange and review of content throughout the translation lifecycle.
Using a translation management system will unburden project managers from using their time to monitor the translation process, and allow reviewers to work more efficiently, focusing only on new and unique content.
-
Translating documents via artificial intelligence/ neural machine translation.
AI translation has evolved to such a degree that it is now suitable to use for many highly regulated document types (like clinical and regulatory materials) when used as part of a hybrid human/machine translation model with qualified post-editor linguists. This workflow reduces translation costs and timelines, while ensuring quality levels are maintained.
Each of these technology types can be a powerful tool in the translation process when used individually, or in combination as an end-to-end workflow. Of all three topics, the one which created the most buzz during the conference and session is the topic of artificial intelligence as it pertains to machine translation. While most companies are familiar with this as a tool, they are skeptical about how it can be used within the pharmaceutical industry.
Before adopting such workflows, many companies dealing with this high-risk content have questions around how quality is measured and ensured, how the technology systems are tested and validated, and how a feedback loop is implemented to ensure continual improvement and learning in the workflow. These are valid questions, and ones that TransPerfect has worked to address in the development process of these technologies.
TransPerfect’s Machine Translation and Artificial Intelligence division has created vigorous processes around “Smart MT” machine translation quality measurement, post-editor linguist training and testing, and client customization to leverage previously translated content and preferred terminology. These engines have been customized and trained specifically for use on pharmaceutical content, with post-editing workflows unique to high-risk pharmaceutical content, ensuring the quality standards that regulators are looking for are met on each project. This technology is being utilized across the pharmaceutical sector and producing promising results in delivering equal or improved to quality to clients with high-risk content, while reducing cycle times and costs associated with translations.
To learn more about how this could be deployed in your organization, contact information@transperfect.com.
