The purpose of this study was to develop a computer-based second opinion diagnostic tool that could read microscope images of lung tissue from resection and lung cells from needle biopsies, and then classify these tissue samples as normal or cancerous. This problem can be partitioned into three areas: segmentation, feature extraction and measurement, and finally classification. One component of this research is to introduce the Mean Shift segmentation algorithm as a prior stage to a kernel-based extension of the Fuzzy C-Means clustering algorithm that provides a coarse initial segmentation. This process is followed by heuristic-based mechanisms to improve the accuracy of the segmentation. The segmented images are then processed to extract and quantify features. Finally, these measured features are used by a Support Vector Machine (SVM) to classify the tissue sample as cancerous or non-cancerous. The performance of this approach was tested using a database of 85 images collected at the Moffitt Cancer Center and Research Institute. These images represent a wide variety of normal lung tissue samples, as well as multiple types of lung cancer.