ABSTRACT:

Risk stratification (characterization) of tumors from radiology images can be more accurate and faster with computer aided diagnosis (CAD) tools. Tumor characterization through such tools can also enable non-invasive cancer staging, prognosis, and foster personalized treatment planning as a part of precision medicine. In this study, we propose both supervised and unsupervised machine learning strategies to improve tumor characterization. Inspired by learning from label proportion (LLP) approaches in computer vision, we propose to use proportion-SVM for characterizing tumors. We also seek the answer to the fundamental question about the goodness of “deep features” for unsupervised tumor classification. We evaluate our proposed supervised and unsupervised learning algorithms on two different tumor diagnosis challenges: lung and pancreas with 1018 CT and 171 MRI scans, respectively, and obtain the state-of-the-art sensitivity and specificity results in both problems.

PROBLEM STATEMENT:

• The main challenging task for machine learning algorithm is to prepare proper dataset for training, we used LIDC-IDRI dataset from Lung Image Database Consortium [32], which is one of the largest publicly available lung nodule dataset. At most four radiologists annotated those lung nodules which have diameters equal to or greater than 3.0 mm.
• The large intra-class variation, especially due to varying shapes of the pancreas is major challenge for detecting them.
• The main problem is considered a complicated process, because of the variability of tumor area of the complexity of determining the tumor position, size, shape and texture.


OBJECTIVE:

Lung cancer is a disease where cells in the lungs multiply uncontrollably. Lung cancer cannot be prevented but its risk can be reduced. So detection of lung cancer at the earliest is crucial for the survival rate of patients. Precise information about the affected area is crucial for the appropriate treatment. The main objective of this paper is to develop an automated and appropriate method for detecting abnormalities in lungs image using artificial intelligence.
The main reason behind the development of this application is to provide proper treatment as soon as possible and protect the human life from lungs cancer.