Early detection and treatment of lung cancer is one of the most effective means to reduce cancer mortality; chest
X-ray radiography has been widely used as a screening examination or health checkup. The new examination
method and the development of computer analysis system allow obtaining respiratory kinetics by the use of flat
panel detector (FPD), which is the expanded method of chest X-ray radiography. Through such changes functional
evaluation of respiratory kinetics in chest has become available. Its introduction into clinical practice is expected in
the future. In this study, we developed the computer analysis algorithm for the purpose of detecting lung nodules
and evaluating quantitative kinetics. Breathing chest radiograph obtained by modified FPD was converted into 4
static images drawing the feature, by sequential temporal subtraction processing, morphologic enhancement
processing, kinetic visualization processing, and lung region detection processing, after the breath synchronization
process utilizing the diaphragmatic analysis of the vector movement. The artificial neural network used to analyze
the density patterns detected the true nodules by analyzing these static images, and drew their kinetic tracks. For the
algorithm performance and the evaluation of clinical effectiveness with 7 normal patients and simulated nodules,
both showed sufficient detecting capability and kinetic imaging function without statistically significant difference.
Our technique can quantitatively evaluate the kinetic range of nodules, and is effective in detecting a nodule on a
breathing chest radiograph. Moreover, the application of this technique is expected to extend computer-aided
diagnosis systems and facilitate the development of an automatic planning system for radiation therapy.