Please use this identifier to cite or link to this item:
|Title:||Pathological diagnosis of bladder cancer by image analysis of hypericin induced fluorescence cystoscopic images||Authors:||Kah, J.C.Y.
Digitized fluorescence endoscopy imaging
|Issue Date:||2005||Citation:||Kah, J.C.Y.,Olivo, M.C.,Lau, W.K.O.,Sheppard, C.J.R. (2005). Pathological diagnosis of bladder cancer by image analysis of hypericin induced fluorescence cystoscopic images. Progress in Biomedical Optics and Imaging - Proceedings of SPIE 5863 : 1-8. ScholarBank@NUS Repository.||Abstract:||Photodynamic diagnosis of bladder carcinoma based on hypericin fluorescence cystoscopy has shown to have a higher degree of sensitivity for the detection of flat bladder carcinoma compared to white light cystoscopy. The potential of the photosensitizer hypericin-induced fluorescence in performing non-invasive optical biopsy to grade bladder cancer in vivo using fluorescence cystoscopic image analysis without surgical resection for tissue biopsy is investigated in this study. The correlation between tissue fluorescence and histopathology of diseased tissue was explored and a diagnostic algorithm based on fluorescence image analysis was developed to classify the bladder cancer without surgical resection for tissue biopsy. Preliminary results suggest a correlation between tissue fluorescence and bladder cancer grade. By combining both the red-to-blue and red-to-green intensity ratios into a 20 scatter plot yields an average sensitivity and specificity of around 70% and 85% respectively for pathological cancer grading of the three different grades of bladder cancer. Therefore, the diagnostic algorithm based on colorimetric intensity ratio analysis of hypericin fluorescence cystoscopic images developed in this preliminary study shows promising potential to optically diagnose and grade bladder cancer in vivo. © 2005 SPIE and OSA.||Source Title:||Progress in Biomedical Optics and Imaging - Proceedings of SPIE||URI:||http://scholarbank.nus.edu.sg/handle/10635/74913||ISSN:||16057422|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
checked on Oct 13, 2019
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.