The vast majority of bladder cancers originate within 600 µm of the tissue surface, making optical coherence tomography (OCT) a potentially powerful tool for recognizing cancers that are not easily visible with current techniques. OCT is a new technology, however, and surgeons are not familiar with the resulting images. Technology able to analyze and provide diagnoses based on OCT images would improve the clinical utility of OCT systems. We present an automated algorithm that uses texture analysis to detect bladder cancer from OCT images. Our algorithm was applied to 182 OCT images of bladder tissue, taken from 68 distinct areas and 21 patients, to classify the images as noncancerous, dysplasia, carcinoma in situ (CIS), or papillary lesions, and to determine tumor invasion. The results, when compared with the corresponding pathology, indicate that the algorithm is effective at differentiating cancerous from noncancerous tissue with a sensitivity of 92% and a specificity of 62%. With further research to improve discrimination between cancer types and recognition of false positives, it may be possible to use OCT to guide endoscopic biopsies toward tissue likely to contain cancer and to avoid unnecessary biopsies of normal tissue.