Background: CADe and CADx systems for the detection and diagnosis of lung cancer\nhave been important areas of research in recent decades. However, these areas are\nbeing worked on separately. CADe systems do not present the radiological characteristics\nof tumors, and CADx systems do not detect nodules and do not have good levels\nof automation. As a result, these systems are not yet widely used in clinical settings.\nMethods: The purpose of this article is to develop a new system for detection and\ndiagnosis of pulmonary nodules on CT images, grouping them into a single system for\nthe identification and characterization of the nodules to improve the level of automation.\nThe article also presents as contributions: the use of Watershed and Histogram\nof oriented Gradients (HOG) techniques for distinguishing the possible nodules from\nother structures and feature extraction for pulmonary nodules, respectively. For the\ndiagnosis, it is based on the likelihood of malignancy allowing more aid in the decision\nmaking by the radiologists. A rule-based classifier and Support Vector Machine (SVM)\nhave been used to eliminate false positives.\nResults: The database used in this research consisted of 420 cases obtained randomly\nfrom LIDC-IDRI. The segmentation method achieved an accuracy of 97 % and\nthe detection system showed a sensitivity of 94.4 % with 7.04 false positives per case.\nDifferent types of nodules (isolated, juxtapleural, juxtavascular and ground-glass) with\ndiameters between 3 mm and 30 mm have been detected. For the diagnosis of malignancy\nour system presented ROC curves with areas of: 0.91 for nodules highly unlikely\nof being malignant, 0.80 for nodules moderately unlikely of being malignant, 0.72 for\nnodules with indeterminate malignancy, 0.67 for nodules moderately suspicious of\nbeing malignant and 0.83 for nodules highly suspicious of being malignant.\nConclusions: From our preliminary results, we believe that our system is promising\nfor clinical applications assisting radiologists in the detection and diagnosis of lung\ncancer.
Loading....