Related Experiment Video
Updated: Oct 10, 2025
05:33
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
Published on: July 11, 2025
376
Electronic case report forms generation from pathology reports by ARGO, automatic record generator for
Gian Maria Zaccaria1, Vito Colella2, Simona Colucci2
1Hematology and Cell Therapy Unit, IRCCS Istituto Tumori 'Giovanni Paolo II', Viale Orazio Flacco, 65, Bari, Italy. g.m.zaccaria@oncologico.bari.it.
Scientific Reports
|December 11, 2021
Summary
An automated tool, ARGO, uses Natural Language Processing (NLP) to convert unstructured pathology reports into structured electronic case report forms (eCRFs) for onco-hematology research. This advances real-world data utilization in cancer research.
Area of Science:
- Oncology
- Hematology
- Medical Informatics
- Natural Language Processing
Background:
- Unstructured real-world (RW) data in onco-hematology presents challenges for research due to limited accessibility.
- Natural Language Processing (NLP) offers a potential solution for standardizing unstructured clinical reports into electronic health records.
Purpose of the Study:
- To develop and validate an automated tool, ARGO, for extracting information from onco-hematology pathology reports.
- To populate electronic case report forms (eCRFs) using NLP, thereby structuring RW data for research.
Main Methods:
- Developed ARGO, an NLP-based tool, to process hemo-lymphopathology reports for diffuse large B-cell, follicular, and mantle cell lymphomas.
- Assessed ARGO's accuracy, precision, recall, and F1-score on internal and external datasets of pathology reports.
Main Results:
- ARGO successfully converted 98.2% of reports into eCRFs.
- High performance was achieved in capturing identification numbers, biopsy dates, specimen types, and diagnoses.
- Metrics demonstrated robust performance across both internal and external validation series.
Conclusions:
- A generalizable NLP tool, ARGO, was developed and validated for generating structured eCRFs from real-world pathology reports.
- This tool enhances the utility of RW data in onco-hematology research by overcoming data structuring limitations.
