A computational software technique that corrects optical tomography aberrations after images have been made could replace hardware-based adaptive optics and bring real-time 3-D microscopic tissue imaging into focus for a broader population of users. Precision is a critical component for correctly diagnosing diseases. Optical aberrations, such as astigmatism or distortion, affect high-resolution tissue imaging, causing areas that should be in focus to become blurry or streaked. Aberrations in imaging can make points appear as slashes or blurs. Graphic image courtesy of Steven Adie. Researchers at the University of Illinois have developed a software technique called computational adaptive optics to replace the hardware-based method, which can focus on only one plane at a time to correct light profiles before they enter the lens. Besides being tedious, this system is bulky and expensive. By using the software program, researchers can correct aberrations after images have been made. This enables them to focus on maximizing the amount of light that their instruments can gather, rather than worrying about minimizing aberration. “Our computational adaptive optics allows one to correct/manipulate the optical data after acquisition, or in real time, and corrections can be made at multiple depths in 3-D,” Stephen A. Boppart, the Bliss Professor of Engineering, told BioPhotonics. “With our approach, we believe we have greater control over corrections, with more degrees of flexibility.” University of Illinois engineers developed a method to computationally correct aberrations in 3-D tissue microscopy. From left, postdoctoral researcher Steven G. Adie, professor P. Scott Carney, graduate students Adeel Ahmad and Benedikt W. Graf, and professor Stephen A. Boppart. Courtesy of L. Brian Stauffer. Boppart’s team demonstrated the technique in gel-based phantoms laced with microparticles, as well as in rat lung tissue. They scanned the samples with an interferometric microscope and collected data with a computer that corrected the images for the entire volume of the sample. “The most highly scattering tissues present the greatest challenge, but we have seen positive results in all tissue types,” Boppart said. “The aberration correction is most obvious in tissue structures that have defined points, edges or boundaries, but we still see improvement in contrast and feature identification in relatively homogenously appearing tissues.” Because the adjustments can be made after or during acquisition, the technique speeds up the process and offers more options for image improvement, he said. “Ultimately, we view this type of computational approach as improving microscopy in general, so that even low-cost microscopes/optical systems that may have many aberrations could all be corrected to give corrected images,” Boppart said. “Now we can focus more on collecting the most signal and data, rather than focusing on collecting the best image (sometimes at the expense of signal).” The researchers now plan to refine the software’s correction algorithms and are implementing this into their clinical systems and studies. They are currently imaging the human eye and expect to show their results within the next six months, he said. “Ongoing studies are implementing our computational adaptive optics technique in our portable intraoperative imaging systems,” Boppart said. “We believe that by generating aberration-corrected images, we’ll be able to better define cells and structures to improve the diagnostic content of our images.” This could result in more informative optical biopsies. “Images are nice to look at, but we need the quantitative information to make objective decisions,” he said. The research, which appeared in the Proceedings of the National Academy of Sciences (doi: 10.1073/pnas.1121193109), was funded in part by the National Institutes of Health and the National Science Foundation.