Sharper microscopic images may be in sight with a new method that uses standard low-light electron multiplying CCD (EMCCD) cameras in a highly unconventional setting to reduce noise and minimize image deterioration. Conventional imaging can translate the image of an object with added electronic noise, causing distortion. The new technique, developed by an electrical engineering team at the University of Texas at Dallas led by professor Raimund Ober, could enhance the accuracy with which quantities of interest — such as the location, size and orientation of an object — are extracted from acquired images. “Our method is about using the EMCCD camera to its fullest potential, beyond what is commonly believed to be possible by the scientific imaging community,” Ober said. “We have figured out through rigorous theoretical developments that when you run an EMCCD camera in such a way that very few photons hit each of its pixels, the resulting image is minimally corrupted by the camera noise.” By increasing the magnification of the image to reduce the number of photons detected in each image pixel, the researchers significantly reduced the camera noise and lessened the deteriorative effect of pixilation. In fact, they attained particle localization accuracy that was twice as high as that obtained with conventional EMCCD imaging. The investigators applied ultrahigh-accuracy imaging modality (UAIM) to the live-cell tracing of a standard protein marker for breast cancer. With the ability to accurately follow the marker’s movement, valuable information on the biology of breast cancer can be gained, the scientists say. “The tracking of individual proteins represents an important way to study cancer and other diseases at the molecular level,” Ober said. “The applications of UAIM for diagnostics and research are promising.” The research, funded by the National Institutes of Health and the Cancer Prevention Research Institute of Texas, was published in Nature Methods (doi: 10.1038/nmeth.2396). For more information, visit: www.utdallas.edu