Diagnosis of Lung Nodule Using Reinforcement Learning and Geometric Measures

This paper uses a set of 3D geometric measures with the purpose of characterizing lung nodules as malignant or benign. Based on a sample of 36 nodules, 29 benign and 7 malignant, these measures are analyzed with a technique for classification and analysis called reforcement learning. We have concluded that this techinique allows good discrimination from benign to malignant nodules. 

• Type: Conference
• Year: 2005
• Where: MLDM
• Authors: Aristófanes Corrêa Silva, Valdeci Ribeiro da Silva Jr., Areolino de Almeida Neto, Anselmo Cardoso de Paiva
• Keywords: Geometric Measures, Lung Nodules, Machine Learning, Malignant Nodules, MLDM 2005