Cancer is having a greater impact on human lives all over the world. It is expected that 22 million new cancer cases will occur by 2030. Lung cancer is the second most common cancer among both genders. Every year, around 1.8 million people are diagnosed with lung cancer. The overall 5-year survival rate of lung cancer is very poor. However, diagnosing lung cancer at an early stage helps to increase the survival rate by 73%. Currently various, imaging modalities are available for medical diagnostics. Among various imaging modalities, positron emission tomography (PET), combined with computed tomography (CT), is used mainly in cancer detection, accurate staging, treatment planning, and monitoring. PET/CT provides better imaging data with the use of anatomic information by CT scan and functional information by PET scan. Computer-aided diagnosis (CAD) systems are developed to assist doctors in analyzing the medical images captured by various imaging modalities in detecting lung cancer. This chapter aims to develop a CAD system for automatic lung cancer detection from PET/CT images using texture and fractal descriptors. The main phases of the developed CAD system are organized as preprocessing, feature extraction, and classification. Wiener filtering and fuzzy enhancement are used in preprocessing to improve image quality. Texture and fractal analysis has been carried out to extract salient image features. Performance of the developed CAD system with various classifiers has been analyzed, and it has been observed that a support vector machine with a radial basis function kernel with width σ = 1 yielded an accuracy of 98.13%.