TI Automate

Save time by automatically coding documents as they get uploaded

- With the introduction of a new Artificial Intelligence service to TI Automate, documents may be automatically classified with their metadata extracted once uploaded to a room. This exciting new feature has the following key capabilities:

• TI Automate is pre-trained on the eTMF reference model, so no additional training is necessary. Your users are up and running with the press of a button.
• In addition to metadata, new investigative sites and new site contacts that are detected in certain documents, such as Form 1572, are automatically added with verification.
• TI Automate has a high accuracy rate that is well above 90% for structured documentation.
• Documents that come in as images are automatically OCR' so that the required metadata may be extracted.
• For non-eTMF rooms, the metadata may be modified and mapped to meet your specific business needs. Site Collaborate and QDMS service offerings are supported for auto-classification, and new document types are easily supported.
• Over 20+ languages are supported out of the box, allowing TI Automate to translate metadata from documentation as needed.
• Just like past TI Automate capabilities, a Confidence Score is assigned to each document and compared against a configurable Confidence Threshold. If the score is high enough, the document may be promoted in the workflow to allow for 'human in the loop' verification.
• TI Automate is a self-training system. Documents that fail to meet the Confidence Threshold are sent for analysis and improvement.