Native fluorescence spectroscopy: An optical tool in delineating oral cancer patients from normal subjects and diabetic patients using urine

• Published in: 2017 Trends in Industrial Measurement and Automation (TIMA) pp. 1 - 4
• Main Authors: Pachaiappan, Rekha, Prakasarao, Aruna, Kesavan, Anjana, Singaravelu, Ganesan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.01.2017
• Subjects: artificial neural network; Artificial neural networks; Biomedical optical imaging; Cancer; Diabetes; Fluorescence; indoxyl sulfate; Lesions; linear discriminant analysis; Native fluorescence spectroscopy; oral cancer; Spectroscopy; urine
• DOI: 10.1109/TIMA.2017.8064822

Many fluorescence spectroscopic studies have proven its novelty and considered as a promising tool in the diagnosis of any health disorders using biofluids. This technique is highly sensitive, real time, non invasive and also it holds many interdependent methods. Urine is one of the most diagnostically important biofluid as it carries specific proteins released from blood used in the collection of significant diagnostic information. In this study, the characterization of urine of normal subjects (14), diabetic (14) and oral cancer patients (14) was carried out using native fluorescence spectroscopy at 280 nm excitation. Thus, obtained fluorescence spectral data were analyzed with the statistical methods linear discriminant analysis (LDA) coupled with artificial neural network (ANN). The results obtained were promising in the classification among normal urine and abnormal urine (diabetic & oral cancer urine).