Enhancement of signal-to-noise ratio for fluorescence endoscope image based on fast digital lock-in algorithm

• Main Authors: Huiquan Wang, Meng Hu, Fang Xia, Meng Guo, Shengzhao Zhang, Zhe Zhao, Guang Han, Jinhai Wang
• Format: Article
• Language: English
• Published: The Royal Society 2021-03-01
• Series: Royal Society Open Science
• Subjects: fluorescence endoscope; tumour; fast digital lock-in algorithm; image snr
• Online Access: https://royalsocietypublishing.org/doi/pdf/10.1098/rsos.200779

In this paper, the signal-to-noise ratios (SNR) of two image channels were enhanced with the fast digital lock-in algorithm. In order to simultaneously improve the quality of white and fluorescence images obtained by fluorescence endoscope, and improve the SNR to achieve a better image processing effect, two sources of white light and near-infrared light of a fluorescence endoscope were modulated, then the acquired images were demodulated into white and fluorescence images. A fluorescent endoscope experimental platform was setup to acquire endoscopic images of a target dyed by indocyanine green. The experimental results showed that the SNR of white and fluorescent images without the lock-in algorithm were 36.56 dB and 33.47 dB, respectively. However, with the lock-in algorithm, the SNR of white and fluorescent images were 39.54 dB and 35.70 dB, respectively. The SNR of white and fluorescent images was increased by 8.2% and 6.7%, respectively, by appling the digital lock-in algorithm. Therefore, this novel fluorescence endoscope based on the fast digital lock-in algorithm can rapidly and simultaneously obtain two-channel images of white light and fluorescence, effectively enhance the SNR of white and fluorescent images, and improve the imaging quality.