New Microscope Captures Large Non-flat, High-Resolution Images in One Snapshot

Duke researchers have created a new microscope capable of capturing extremely large and detailed images of objects with uneven or curved surfaces in a single snapshot. This technology could accelerate progress in medical diagnostics, research, and industrial quality control by eliminating traditional limitations tied to sample flatness and image acquisition time. 

“Although traditional microscopes assume the sample is perfectly flat, real-life samples such as tissue sections, plant samples or flexible materials may be curved, tilted or uneven,” explained Roarke Horstmeyer, senior author of the paper published in Optics Letters. Their new microscope, named PANORAMA, adjusts the focus dynamically across the sample surface so that every part remains sharply in view without the need for slow scanning or costly specialized lenses.

The PANORAMA microscope captures submicron details—ranging from 1/60 to 1/120 the diameter of a human hair—over an area about the size of a U.S. dime in a single exposure. It produces gigapixel-scale images that contain 10 to 50 times more pixels than a typical smartphone camera photo.

According to co-author Haitao Che, the system is designed for applications requiring large-area, detailed imaging. “For instance, in medical pathology, it could scan entire tissue slides, such as those from a biopsy, at cellular resolution almost instantly.”

The microscope overcomes the usual trade-off of conventional microscopes, which must choose between small areas with high detail or large areas with low detail. Typically, capturing high-resolution gigapixel images requires complex optics or time-intensive tile scanning, with additional focusing adjustments necessary because real samples are rarely perfectly flat at centimeter scale.

The PANORAMA system works like a multi-camera array combining a telecentric lens originally designed for chip inspection with a large tube lens projecting onto an array of 48 small cameras. Each camera independently focuses on a portion of the sample, allowing the entire curved or uneven surface to stay in focus simultaneously. This removes the need for mechanical scanning, reducing image capture from up to an hour to a single snapshot, with automatic image stitching taking about 5 to 10 minutes.

Testing involved imaging a rat brain tissue slide, producing a 630-megapixel brightfield image in one shot that clearly revealed cellular structures down to 0.84 microns. The team also imaged onion skin laid over a curved surface using brightfield and fluorescence simultaneously, with sharp focus on the curved layer and clear visualization of cell walls and stained nuclei. 

 “We saw a huge jump in throughput and flexibility: no more moving parts, no tedious focus-stacking, and no blind spots between cameras,” Horstmeyer summarized. The team is now advancing this technology by adding more cameras or larger sensors to capture even larger fields of view, developing automated focus adjustment, and exploring computational methods to enable 3D reconstructions and live microscopic imaging.