Project

Optimization of low dose CT for lung cancer screening: finding the best balance between radiation risk and performance in terms of image quality and success rate of computer aided detection

Code
3G0B1922
Duration
01 January 2022 → 31 December 2023
Funding
Research Foundation - Flanders (FWO)
Research disciplines
  • Natural sciences
    • Other physical sciences not elsewhere classified
  • Medical and health sciences
    • Diagnostic radiology
    • Medical imaging and therapy not elsewhere classified
    • Quality assurance
Keywords
texture analysis and radiomics radiation risk computer aided detection
 
Project description

Lung cancer causes a large mortality, worldwide and in Flanders,

with 3822 cases in 2016. The only technique with proven mortality

reduction is lung cancer screening (LKS) with CT scans. In 2020, a

mortality reduction of 26% (men) and 41% (women) was measured in

the Dutch-Belgian "NELSON" study. The Flemish 'Task Force Lung

Cancer Screening' therefore brings together experts to design and

subsequently support an LKS program for population groups where

this improvement can be continued. All (expected) questions from the

responsible Flemish government must be answered. With this

project, radiologists and physicists from the Task Force want to find

adequate answers to the following questions: what is the radiation

risk associated with the CT scans? Can imaging be optimized for

even lower X-ray doses? How are computer algorithms for detecting

and measuring cancers performing?

The greatest challenge is to develop new test methods in a world

with rapidly changing technology - e.g. artificial intelligence. Three

university groups will combine their expertise with (1) risk

assessment, (2) evaluation of clinical image quality and (3)

computer-aided detection and characterization of cancer. All new

techniques and test criteria will be bundled in a unique test protocol

to assess and optimize the quality of each CT protocol or computer

algorithm for LKS.