Kavitha Ganesan. Sequential Decision Making to Improve Lung Cancer Screening Performance Using QCD Algorithm
Neural Parallel and Scientific Computations 29 (2021), No. 3, 173-184

https://doi.org/10.46719/npsc20212934

ABSTRACT.   
Area of segmentation and lung cancer (large) is a non-trivial problem. We are, for the segmentation of the high density of lung lesions, and proposes a new method of such fully automated. Our approach involves two major steps. Low-dose computed tomography (of LDCT) conventional method limits the processing speed. Lung-shaped face cannot explain. According to the missing data. In an adaptive combination of denser mesh vertices and search profile is required. This method is substantially are used to match the contour segments of the lungs. The initial position of the rib cage of Fast Convolution Diagnosis FCD was determined by the detection method of the device. Secondly, the best way is, furthermore, are used for surface to adapt the initial segmentation result to the lungs. The left and right lung is divided individually. It recommends ways to improve processing speed. Lung shape encounter can be explained. Security and authorization based on the data. In the low-dense mesh adaptation apex combined search profile is required. 30 40 Evaluation of abnormal data set (lung) and 20 normal left / right lungs respectively caused 0.975 +0.23 0.84 + 0.0006 mm, and the average absolute error from the surface, the average coefficient of dice. In the same experiment data set 30, our method is compared with two commercially available lung division, it has been shown to provide a better segmentation statistically significant results. In addition, our methods, and generally be applied to the FCD, it will be applied to a large shape model.

Keywords: Sequential Decision, Lung-shaped face, Convolution Diagnosis.