RadPhysics Services LLC is proud to be a co-author of Automated Error Labeling in Radiation Oncology via Statistical Natural Language Processing. Recently published by MDPI (Multidisciplinary Digital Publishing Institute), the peer-review article describes how natural language processing (NLP)-aided statistical algorithms have the potential to significantly improve the discovery and reporting of medical errors in radiation oncology. We believe such an NLP-aided system can help relieve human reporters of the burden of event type categorization. In our work, we created an automated, streamlined prototype model for error incidents. Further, we demonstrated how text-classification models developed with clinical data from a full service radiation oncology center (test center) can predict the broad level and first level category of an error given a free-text description of the error. In our work, all but one of the resulting models had an excellent performance as quantified by several metrics. The results also suggest that more development and more extensive training data would further improve future results.
MDPI a leading publisher of open access scientific journals. A pioneer in scholarly, open access publishing, MDPI is among the largest publishers in the world in terms of journal article output, and is the largest publisher of open access articles.
