IOMED will participate in IberLEF as the organizing entity for an NLP (Natural Language Processing) task in the upcoming month of September.
At IOMED, we believe that the current advanced state of clinical NLP in Spanish presents an opportunity for the development of automated tools for identifying sections in clinical documents. As such, the tech team has worked in collaboration with the research group at HITZ (Basque Center for Language Technology) to generate an annotated corpus of medical sections.
The task to be carried out is novel, as it is the first time that the detection and classification of text portions corresponding to each section in a clinical document are being proposed for Spanish.
The project, which will be presented at IberLEF, involves identifying seven sections that could appear in a document. The chosen sections are tailored to current specific needs and include: current condition, medical history, family history, examination, evolution, treatment, and finally, the section derived from/to, which contains information about another hospital or service where part of the current condition was treated.
IOMED anticipates that this will be of interest both to the wide potential array of research groups working in medical NLP and to industrial companies, given that the current advanced state of clinical NLP in Spanish presents an opportunity for the development of these types of tools that can offer a competitive advantage over other medical NLP systems.