Information Extraction From Blood Test PDFs: Systematic Review and Case Study

Automated extraction of information from files is increasingly essential due to the growing volume of data. However, extracting information from blood test results remains a relatively unexplored area. Therefore, the objective of this paper is to present a system that is capable of automatically extracting blood test variables. To this end, this study proposes a comprehensive review of the methodologies and tools employed in extracting information from blood test PDFs based on a systematic review and a case study that presents the developed system in detail in all its stages, in order to validate the proposed system. The proposed methodology is validated based on the use of precision, recall and F1 Score evaluation metrics. To the best of our knowledge, there is no existing literature that conducts a study analogous to the one presented and validated in this paper through the provided case study.