Abstract:
The aim of the study is to estimate and identify the GTV of NSCLC with fluorine-
18 fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) scan instead
of using the CT imaging, The PET scan so useful in estimation of sub mucosal
extension of the tumor as well as in staging and restaging of lung carcinomas and
changing in gross target volume either positively (+) or negatively (-).
This study was conducted in Kuwait Cancer Control Center (kccc) in Kuwait City,
during the period 2016 to 2018.
The study sample included 156 patients with non- small cell lung carcinoma
underwent whole body PET/CT scan.
The features extracted from PET/CT images using Second order statistic and All
these features were calculated for all images and then the data were ready for
discrimination which was performed using step-wise technique in order to select the
most significant feature that can be used to classify the lung cells from PET/CT
images and the results show that The classification showed that the lung cells were
classified well from the rest of the tissues although it has characteristics mostly
similar to surrounding tissue. The classification score matrix generated by linear
discriminate analysis and the overall classification accuracy of lung cells 96.0%,
were the classification accuracy of cardiac 91.6%, lung accuracy 100%, the tumor
99.6%, While the submucosal showed a classification accuracy of 91.2%.
Texture analysis depending on the relative attenuation coefficient of tissues and can
used to avoid invasive technique if the base line for individual tissues being
determined and algorithmic aided computer have been applied.
