TechLab: Galenos.NLP

The main articles are:

• Speech Recognition (Technology, Market)
• Speech technology: On the path from recognition to understanding

2025: Presentation of Galenos.NLP. Registration by Rospatent

TechLAB, a developer of digital health services, presented a solution based on artificial intelligence on January 16, 2025. An AI service called Galenos.NLP, designed to extract clinical parameters from unstructured medical text, is registered in the Register of Computer Programs (Certificate No. 2024683750), which is supervised by Rospatent.

This service is already the second solution in the company's portfolio using artificial intelligence, focused on analyzing text medical data. It is based on Natural Language Processing (NLP) technology, which helps to recognize human speech, interpret, process and generate it.

Despite the fact that in recent years medical organizations have been actively using structured electronic medical documents (consultation protocols, discharge epicrises, etc.), a significant part of important medical information still remains presented only in an unstructured form, for example, in the text of the doctor's conclusion. It gets there naturally: sometimes it is more convenient and faster for a doctor to write down an important thing in a single text, and not to distribute information across the proposed structured fields to present the entire clinical picture of the disease. Finally, many electronic medical documents from an earlier period exist exclusively in textual form (including in the form of scanned paper documents).

When working with such medical text, you cannot do without digital tools for processing the natural language. Thanks to NLP technology, the service from TechLAB provides streaming extraction of the necessary indicators, fills them with the database of the information system and generates reports and selections in the usual interface.

For example, using the Galenos.NLP service, it is possible to determine which patients with a left ventricular ejection fraction of less than 40% have preferential drug provision available, this information is almost always contained in text rather than structured fields, and the AI based on Galenos.NLP is able to identify and analyze such information.