New microscope sees inside tissue in three dimensions

A new technique to see inside tissues in three dimensions so doctors can make better and much faster diagnoses, including even during an operation, has been invented by scientists in the US. In the new system a flat sheet of light is used to illuminate a series of thin slices through a block of tissue. The light that emerges is captured by a camera, which builds a three-dimensional picture of the entire specimen. Chris Smith spoke to Jonathan Liu from the University of Washington who helped to invent the technique…

Jonathan - The way to think of it would be as a flatbed scanner for tissues. The sample sits on top of a glass plate; all of the optics are underneath so the light and all the complexity is hidden underneath that glass plate, all the user has to do is place the tissue on top.

Chris - What sort of tissue specimens: how big, what can you image?

Jonathan - That’s something that’s rather unique about our system in that it’s somewhat unconstrained in terms of the types of specimens that you can place on top. We’ve imaged tissues as large as 5 x 5 cm; these are relatively large surgical excisions. We can also image smaller specimens 1 mm in diameter core needle biopsies that are obtained from patients who are suspected to have tumours.

Chris - So if you had a person undergoing surgery, the key question a surgeon wants to be sure of when they’re operating is “have I removed all of this person’s cancer”, for example? You could take the tissue that’s come out of the patient and you could image the block and see if there are what we call ‘clear margins’  - there’s an area around the tissue where there are not cancerous deposits so the surgeon knows that they haven’t got to return that person to the theatre later for another operation?

Jonathan - That’s correct. There are alternative technologies that have been attempted. For example, frozen sectioning where they freeze the tissues so that they can cut the tissue very rapidly during surgery, but these techniques generally do not produce very reliable results and for certain tissue types. For example, fatty breast tissues, they don’t work well because fat does not freeze well so it’s very difficult to obtain a high quality image.

Chris - So tell us then how it actually works. You get some fresh tissue hot out of the patient, it goes on your glass surface - what’s going on under the hood to make this possible?

Jonathan - Traditionally with pathology the tissue has to be chemically processed, mounted in a wax block, sliced into very thin sections that are mounted on a glass slide and looked at under a traditional microscope. With our technology we don’t have to cut the tissue, we use light to slice into the tissue. This is something we call optical sectioning as opposed to physical sectioning with a knife.

So we send in a thin sheet of light, and we image that sheet with a camera so that we can see an image that looks like the tissue has been sliced into a very thin section without having to actually cut into the tissue.

Chris - I’ve got my block of tissue sitting on top of the microscope, the lights coming in from below. Does it come in at an angle to create that light sheet and then how does the camera see what the light sheet is seeing?

Jonathan - That’s correct. In order to image the light sheet, our camera has to be situated at a 90 degree angle to that light sheet as it enters the tissue. So instead of sending in the light perpendicular to the surface of the tissue, we send it in at a 45 degree angle, and then the output beam also exits the tissue surface at a 45 degree angle. As a result we can image these oblique light sheets that are cutting into the tissue at 45 degrees. As we scan the tissue we collect a series of these oblique light sheets so that we can obtain a three dimensional volume of the tissue.

Chris - Can your computer recompile each of your sheets on slices of the tissue to produce a 3D model effectively on the screen of what the microscope is seeing?

Jonathan - Yes, exactly. That’s the intent of the device to collect these 2D light sheets and to reconstruct them into a 3D volume so that we can display to the pathologists and other clinicians the 3D microarchitecture of the tissue, which should allow them to understand the tissue, understand the disease, and to guide patients treatments ultimately more accurately. So we feel, and we’ve shown in the paper, that there are much more accurate diagnoses that can be made based on a 3D information.

Chris - If this lives up to your expectations, what sort of a difference will it make for the patient?

Jonathan - For treatment, this can make a huge difference. There is a problem that’s recognised now, especially for prostate cancer patients as well as breast cancer patients, a lot of these patients are being overtreated where they’ll receive surgery, or chemotherapy, or radiation therapy when it’s not needed and these therapies all have side effects. So it’s very important that we can stratify the patients to determine which patients should be treated, and which patients should maybe undergo active surveillance and perhaps the disease won’t actually be very malignant.

In prostate cancer, most of the cancers are not very aggressive but, for a small fraction of patients, they can lead to death and we need to be able to identify those patients for the appropriate treatments.