ngenuity 3d visualization system  (alcon inc)

Journal: Scientific Reports

Article Title: Digital image enhancement using deep learning algorithm in 3D heads-up vitreoretinal surgery

doi: 10.1038/s41598-025-98801-7

Figure Lengend Snippet: Schematic diagram of digital image enhancement using deep learning algorithm. The vitreoretinal surgeon manually adjusted eight image parameters (brightness, saturation, contrast, hue, gamma, cyan, magenta, and yellow) to optimize the original surgical image and placed the adjusted image and parameter values in the algorithm architecture to train the deep learning model. After training the deep learning model, eight image parameter values for optimization were predicted when the original surgical image was applied to the algorithm. By inputting the predicted parameter values into the Ngenuity 3D Visualization System (Alcon Laboratories, Fort Worth, TX) capable of digital image enhancement, an optimized surgical image was finally obtained.

Article Snippet: By inputting the predicted parameter values into the Ngenuity 3D Visualization System (Alcon Laboratories, Fort Worth, TX) capable of digital image enhancement, an optimized surgical image was finally obtained.

Techniques:

Journal: Scientific Reports

Article Title: Digital image enhancement using deep learning algorithm in 3D heads-up vitreoretinal surgery

doi: 10.1038/s41598-025-98801-7

Figure Lengend Snippet: In-vitro experiments for evaluation of optimized fundus image using Ngenuity 3D visualization system. ( A , B ) 121 high-resolution fundus images of epiretinal membrane (ERM) obtained from Google images and a dataset. The fundus images of ERM obtained were directly observed in a 3D heads-up surgery setting. ( C ) Manual manipulation of scroll bars within the system. An optimized fundus image was obtained by adjusting the image parameter values. ( D – F ) Pelli–Robson contrast sensitivity chart. This chart was printed in a small size and directly observed in the 3D heads-up surgery setting by 25 investigators. Subsequently, it was optimized, and the number of letters that can be read before and after optimization was compared.

Article Snippet: By inputting the predicted parameter values into the Ngenuity 3D Visualization System (Alcon Laboratories, Fort Worth, TX) capable of digital image enhancement, an optimized surgical image was finally obtained.

Techniques: In Vitro, Membrane

Journal: Scientific Reports

Article Title: Digital image enhancement using deep learning algorithm in 3D heads-up vitreoretinal surgery

doi: 10.1038/s41598-025-98801-7

Figure Lengend Snippet: Measurements of color contrast ratio (CCR) and epiretinal membrane (ERM) area in experimental fundus images on Ngenuity 3D visualization system. ( A , B ) On the original fundus image of ERM, the CCR was measured by comparing the red, green, and blue values between the darkest pixel (number 1 on C) and the brightest pixel (number 2 on C) on the prominent retinal folds. Based on this original fundus image, the CCR was measured to be 1.17. ( C , D ) The optimized fundus image was measured in the same manner as the original fundus image, and the CCR value calculated using this method was 2.12. This value is higher than that of the original fundus image, which implies that the folds of the ERM can be observed more clearly; in other words, visibility is improved on the optimized fundus image. ( E , F ) The same ERM was measured to be 19,777 and 22,581 pixels in the original and optimized fundus images, respectively, although the optimized fundus image was measured more extensively.

Article Snippet: By inputting the predicted parameter values into the Ngenuity 3D Visualization System (Alcon Laboratories, Fort Worth, TX) capable of digital image enhancement, an optimized surgical image was finally obtained.

Techniques: Membrane