Autonomous Detection and Classification of Pulmonary Nodules with Mobile Deployment

Umapujitha S
2018 International Journal for Research in Applied Science and Engineering Technology  
Pulmonary nodules are round oval shape growth found in the lungs. These growths may or may not be cancerous in nature. False detection of these nodules has severe effect on the patient. Detection of these nodules in the early stage can increase the rate of survival. Nowadays CT (Computer Tomography) scan is widely used for imaging to X-Ray images and cost efficient when compared to MRI (Magnetic Resonance Imaging) and PET (Positron Emission Tomography). However, the problem merges due to time
more » ... erges due to time constraints in detection of these modules and human error classifying them as malignant or benign. Hence pulmonary nodule detection system uses image processing technique to detect the nodule present in the CT scan image. The image processing procedure includes image acquisition, preprocessing, segmentation, and feature extraction. The machine learning helps in classifying the nodules as malignant or benign. After classification of the nodules all the results from the previous steps and it is uploaded in the server. These data can be accessed by the doctors globally with proper authentication. The aim is to improve the system accuracy by different segmentation and classifiers.
doi:10.22214/ijraset.2018.6288 fatcat:gldugked7vf57ccy6wsgjrgx6i