New Surgical Microscope Offers Precise 3D Imaging Using 48 Tiny Cameras

While these microscopes provide a sense of three-dimensionality, they do not offer precise measurements, making it difficult to estimate exact distances or shapes in complex environments, especially when lighting is uneven or tools obstruct the view.

These challenges have slowed advancements in surgical automation and real-time feedback systems. Additionally, pre-operative 3D scans, like MRIs or CTs, do not update during surgery as tissues shift. Optical coherence tomography (OCT) can offer real-time data but is limited in scope and difficult to interpret. To address these challenges, researchers have developed a new surgical microscope that provides precise 3D measurements in real-time during surgery.

The new solution, known as the Fourier lightfield multiview stereoscope (FiLM-Scope), was developed by researchers at Duke University (Durham, NC, US) and deploys 48 tiny cameras arranged in a grid, all focused through a single high-throughput lens. Each camera captures a high-resolution image from a slightly different angle, producing 48 images of the same scene with 12.5 megapixels each. The field of view is large—about 28 by 37 millimeters—and offers fine detail down to 22 microns.

The system processes these multiple perspectives using a specially designed self-supervised algorithm to create a 3D map of the scene in real-time. The algorithm doesn’t need pre-existing data or models and can reconstruct surface shapes with an accuracy of 11 microns over a depth range of one centimeter. The FiLM-Scope was tested for its performance and capability to create real-time 3D maps, which can digitally zoom or shift the view without moving the microscope.

The study, published in Advanced Photonics Nexus, confirmed that the FiLM-Scope could improve the precision and efficiency of surgery by providing detailed 3D imaging during live procedures. The system’s ability to convert standard images into precise 3D measurements opens new possibilities for both manual and robotic microsurgery. This technology could also have applications in other fields requiring high-accuracy 3D visualization, such as materials science and microfabrication. 

Related Links:
Duke University