Virtual Biopsy Allows Non-invasive Skin Analysis Using Lasers, Creates High-Resolution Reconstruction of Tissue Samples

Researchers have developed a new technique that uses lasers to penetrate tissue to create a high-resolution, 3D reconstruction of cells under study. The virtual reconstruction allows the generation of cross-sectional images that mimic those created by a standard biopsy.

Standard Procedures in Histopathology

When dermatologists or surgeons take a biopsy from a person's body, the tissue is sent to a pathologist, who slices the sample into thin layers. Each layer is stained with chemicals known as hematoxylin and eosin (H&E) to allow the pathologist to observe the patterns, shapes, and structures of cells easily.

H&E slides have long been the gold standard for pathologic detection of cancer and tumor margin detection. It offers a view of the tissue down to the micron scale and has been a routine part of diagnostic procedures for both cancer and noncancer pathologies.

However, producing H&E tissue sections is not only labor-intensive but also invasive and time-consuming. This method is also irreversible, meaning once a biopsy is sliced in one direction, it cannot be sliced in another way to offer a different view.

Noninvasive Approach to Skin Analysis

At Stanford University Medical Center, experts have created a new technique to replace gold-standard pathology slides for disease diagnosis. They improved the resolution of these scans to allow them to pick up information that would be extremely hard to observe otherwise.

The method was developed by Yonatan Winetraub, an Adam de la Zerda graduate student. For almost a decade, de la Zerda and his colleagues have explored a different approach to seeing the human body's interior using optical coherence tomography (OCT) . Ophthalmologists typically use OCT scans to image the back of the eye by measuring how light waves from a laser bounce off a tissue to create a rendering of its internal structure.

Winetraub and de la Zerda enhanced the OCT scans to work in organs other than the eye. They developed new data collection hardware and new processing methods to allow them to see smaller and smaller tissue details. They realized the OCT images they created were getting more minor to the H&Es regarding what they could show.

The higher resolution of the OCT images paved the way for a new method of diagnosing diseases without producing H&Es. To make the picture look familiar, the research team turned to artificial intelligence to help convert OCT scans into flat images that resemble H&E slides.

According to Winetraub, the uniqueness of their work lies in the method they developed to align OCT and H&E image pairs. This enables machine-learning algorithms to train on absolute tissue sections and provide clinicians with more accurate virtual biopsies.

The new method can be helpful for noninvasively scanning the skin for unhealthy cells and providing rapid results on biopsies taken in any part of the body. Additionally, it can offer more information than the current diagnostic techniques cannot. Although more work is needed to move the approach toward these applications, the team is confident that their technique will provide clinicians with a new way to conduct biopsies.

Check out more news and information on Biopsy in Science Times.