2D PANORAMAS FROM CYSTOSCOPIC IMAGE SEQUENCES AND POTENTIAL APPLICATION TO FLUORESCENCE IMAGING

This work describes an algorithm for the automatic construction of 2-D panoramas from sequences of images taken from a cystoscopic exam. During a cystoscopy, the existent distance between the cystoscope and the bladder walls allows the clinician to observe only a reduced region of the zone of interest. A panoramic representation of this zone could represent a better visual support to the clinician in the localization of eventual lesions and in the application of further exams. We also explore the application of this algorithm to endoscopic fluorescence imaging, one of the main procedures used in the detection of epithelial cancers.