Translator Disclaimer
28 February 2012 Intercomparison of EMCCD- and sCMOS-based imaging spectrometers for biomedical applications in low-light conditions
Author Affiliations +
Hyperspectral imaging provides means for characterizing large biological samples with microscopic spatial resolution and a narrow spectral sampling interval. However, this approach requires having a measurable light signal in each spectral band. Overcoming the limitations imposed by working with biological samples requires the use of a highly sensitive sensor to detect weak signals. For this study we have built and compared the performance of two imaging spectrometers using optimized for low light environments: an electron-multiplying CCD (EMCCD) and a scientific CMOS (sCMOS). Both systems have been designed to lower the risk of damaging photosensitive samples, delay the bleaching of fluorophores and detect weak fluorescence signals. The cameras work within the VNIR spectral region (400 nm - 900 nm) with a spectral sampling lower than 4 nm. The produced images have scene pixel sizes smaller than 25 μm and a field of view larger than 25 mm. The systems have been tested side to side measuring the diffusion front of a fluorescent tag in samples of porcine skin in challenging light conditions. The study aimed to show the advantages and limitations of each approach. Preliminary results show good performance of the EMCCD for fluorescence applications, whereas more experimental results are needed to be able to conclude on the performance of the sCMOS sensor. However, the sCMOS appears promising for imaging scenes with high dynamics in low light settings.
© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
J. Hernandez-Palacios and L. L. Randeberg "Intercomparison of EMCCD- and sCMOS-based imaging spectrometers for biomedical applications in low-light conditions", Proc. SPIE 8215, Design and Quality for Biomedical Technologies V, 82150Q (28 February 2012);

Back to Top