Background <p>Innovative imaging technologies combined with artificial intelligence (AI) and endoscopy have the potential to increase the detection rate of mucosal lesions and better characterize tumor biopsies.</p> Objectives <p>Identification of pathological changes during endoscopy using AI algorithms and evaluation of intact and unstained endoscopic biopsies using high-resolution imaging techniques to obtain new three-dimensional (3D) information about the complex architecture and structural composition of tumors.</p> Materials and methods <p>Developments and challenges in the mentioned technological fields, assessment of the potential application of new imaging biomarkers in combination with AI in tumor diagnostics.</p> Results <p>Using AI, mucosal changes can be automatically detected during endoscopic examinations and characterized in real time. Deep learning networks predict biomarkers of tumor diseases for classification in standard histopathological tissue sections. New 3D imaging methods on unstained and intact tissue samples promise to provide additional information to data from routinely performed morphological and molecular analyses.</p> Conclusion <p>AI algorithms in conjunction with endoscopy, 2D histopathological data, and nondestructive 3D examinations of intact, unstained tumor biopsies have the potential to revolutionize diagnosis, treatment and research in the field of cancer. The extent to which AI-based imaging biomarkers together with data from routine examinations help to classify tumors more comprehensively, especially in the age of precision medicine, and to optimize treatment results through the selection of the best possible therapy options needs to be evaluated in larger clinical studies.</p>