A section identification tool: Towards HL7 CDA/CCR standardization in Spanish discharge summaries

•We present a system for Section Identification in medical texts written in Spanish.•The system follows the standard HL7 CDA/CCR.•We have defined an annotation model for Electronic Discharge Summaries (EDS).•We have applied it to annotate 300 EDSs, with a high Inter Annotator Agreement.•We evaluate the contribution of different Rule-Based and Machine Learning approaches.•Language Model tuning is a key factor as transfer learning provides the best results.•We also study the generalization ability of models trained in different hospitals.