Integrating advanced NLP techniques with the OMOP CDM enhances the accuracy and completeness of hematologic oncology data
Accurate data capture in hematologic oncology is crucial for understanding disease progression and outcomes, but traditional methods often miss complex variables. This study applies a natural language processing (NLP) pipeline within the OMOP-CDM to enhance the capture of key variables in diseases like chronic lymphocytic leukemia (CLL) and multiple myeloma (MM), improving the quality and depth of oncology research data.
Download the poster (as showcased at the OHDSI UK 2024 conference) by completing the form.
The abstract also earned us a spot for a lightening presentation, presented by Dries Hens (Chief Medical Officer, co-founder).
Authors: Clara L. Oeste, Iege Bassez, Jana Van Canneyt, Alina Kramchaninova, Lucas Sterckx, Narges Farokhshad, Shahbaz Pervaiz, Geert Van Gorp, Dries Hens
Background
Accurate data capture in hematologic oncology is crucial for understanding disease progression and outcomes, but traditional methods often miss complex variables. This study applies a natural language processing (NLP) pipeline within the OMOP CDM to enhance the capture of key variables in diseases like chronic lymphocytic leukemia (CLL) and multiple myeloma (MM), improving the quality and depth of oncology research data.
Conclusion
We demonstrate the effectiveness of integrating an NLP pipeline with structured data mining in OMOP CDM databases to improve data capture in hematologic oncology. High precision, recall, and F1 scores validate the reliability of this approach, which enhances the quality of datasets and supports large-scale, multicenter research. Our findings highlight the potential of NLP to significantly improve data management and insights in cancer research.
Complete the form above to download the poster and discover the methods used and the results obtained.
Background
Accurate data capture in hematologic oncology is crucial for understanding disease progression and outcomes, but traditional methods often miss complex variables. This study applies a natural language processing (NLP) pipeline within the OMOP CDM to enhance the capture of key variables in diseases like chronic lymphocytic leukemia (CLL) and multiple myeloma (MM), improving the quality and depth of oncology research data.
Conclusion
We demonstrate the effectiveness of integrating an NLP pipeline with structured data mining in OMOP CDM databases to improve data capture in hematologic oncology. High precision, recall, and F1 scores validate the reliability of this approach, which enhances the quality of datasets and supports large-scale, multicenter research. Our findings highlight the potential of NLP to significantly improve data management and insights in cancer research.
Complete the form above to download the poster and discover the methods used and the results obtained.
