Effective And Reliable Lung Segmentation Of Chest Images With Medical Image Processing And Machine Learning Approaches
- Resource Type
- Conference
- Authors
- Chanda, Pramit Brata; Sarkar, Subir Kumar
- Source
- 2020 IEEE International Conference on Advent Trends in Multidisciplinary Research and Innovation (ICATMRI) Advent Trends in Multidisciplinary Research and Innovation (ICATMRI), 2020 IEEE International Conference on. :1-6 Dec, 2020
- Subject
- Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Engineered Materials, Dielectrics and Plasmas
Engineering Profession
Fields, Waves and Electromagnetics
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Image segmentation
Pulmonary diseases
Computed tomography
Lung
Lung cancer
X-ray imaging
Diseases
CAD
X-Ray
preprocessing
Adaptive
segmentation
Threshold
PSNR
Sensitivity
Edge
Precision
SSIM
- Language
Today Identification and prediction Of lung disease is become very crucial and challenging work. The image processing and machine learning based approaches are very much commonly used methodology to detection of lung cancer appropriately. Detection of lung diseases can improve the mortality rate or chances of recovery rate of people become higher. However identification and segmenting of lung cancer in a particular image is quite harder task. The methods can apply on CT and MRI based images for identify the disease more effectively. Pattern recognition is basically used to recognize more perfectly the disease affected region. Doctors may take decision about the disease according to the CT or MRI based scan report. The abnormal images are used to segment the portion of cancer affected area with Otsu threshold based segmentation. Here the adaptive threshold methods are used for segmentation for it can changes the threshold parameters dynamically. The gradient or sobel based detection is applied to detect all the edges in lung area accurately. According to image quality parameters measurement the value decisions can be taken. Here noise removal can be done using Gabor filtering. The algorithm of adaptive threshold based denoising approach provides more than 90 per cent of the accuracy rates than other approach .