Lung cancer classification on radiology reports using Natural language processing
Principal Investigator
Name
Cyrille KESIKU
Degrees
MSc
Institution
University of Deusto
Position Title
Ph.D. Student
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-1124
Initial CDAS Request Approval
Sep 19, 2023
Title
Lung cancer classification on radiology reports using Natural language processing
Summary
Lung cancer generally has a poor prognosis, with an overall 5-year survival rate of 20.5%. However, detecting lung cancer at an early stage has a better prognosis and is more amenable to treatment. The U.S. Preventive Services Task Force (USPSTF) recommends annual lung cancer screening with LDCT for individuals with a smoking history of 20 pack-years or more, who currently smoke or have stopped smoking within the last 15 years, and are between 50 and 80 years old. The natural language processing model may help facilitate lung cancer screening from biomedical patient reports. Several databases can be used for validation to compare the model's effectiveness. In this project, we study the capacity of NLP models in lung cancer research based on patients' biomedical (Radiological and diagnostic) reports.
Aims
The aim of this project is to investigate the capability of NLP models in lung cancer research by exploiting several databases for the validation of the new models created. This project is part of a doctoral research case study in biomedical engineering and Artificial Intelligence.
Collaborators
Begonya Garcia-Zapirain Soto (Professor. Supervisor)