UCLA-developed new “virtual histology” technology may reduce need for skin biopsies

Instead of surgically removing a piece of skin and submitting it to a lab and waiting for several days for results, your dermatologist snaps photographs of the suspicious lesion and then quickly creates a clear, microscopic image of the skin.

This could become routine in clinics, the result of a revolutionary “virtual histology” technology being developed by researchers at the UCLA Samueli School of Engineering and the David Geffen School of Medicine at UCLA according to a recent article in Light: Science & Applications, a journal of the Springer Nature Group. Histology is the study of the microscopic structure of tissues.

“This process bypasses several typical steps used in diagnosis -; including skin biopsy, tissue fixation, processing, sectioning and histochemical staining. The images look like sections of biopsied, stained histochemically-stained skin that has been taken from microscope slides,” Aydogan Ozcan (Chancellor’s Professor and Volgenau chair for Engineering Innovation at UCLA Samueli) was the study’s lead writer.

The technology is in development and research for more than three years, may provide a new avenue for rapid diagnosis of malignant skin tumors, reducing the need for of unnecessary skin biopsies and allowing earlier diagnosis of skin cancer. The technology was previously restricted to slides of untreated tissue obtained via a biopsy. This report is the first to apply virtual histology to intact tissue that is not biopsied.

“The current method of diagnosing skin diseases such as skin cancer is based on an invasive biopsy and histopathological analysis. For patients, this often results in unnecessary biopsies, scars as well as multiple visits to doctors. It also can be expensive for patients as well as the health care system,” said Dr. Philip Scumpia, assistant professor of dermatology and dermatopathology at the David Geffen School of Medicine at UCLA and the West Los Angeles Veterans Affairs Hospital and an affiliate of the UCLA Jonsson Comprehensive Cancer Center. “Our method may provide a biopsy-free option, offering photographs of skin structure at resolution that is cellular-level.”

Ozcan Scumpia, Ozcan, and Dr. Gennady Rubystein, a dermatologist at Dermatology and Laser Centre in Los Angeles, developed a deep-learning framework that transforms images of intact skin taken by an emerging non-invasive optical technology known as reflectance confocal microscope (RCM) into a format that is easy to use by dermatologists as well as pathologists. RCM images require special training since they are black and white and, unlike conventional histology (which includes nuclear features of cells), require special training.

“I was surprised to see how simple it is to use this technology to transform the images into the ones I usually see of skin biopsies which are processed with traditional chemical fixation and tissue staining under microscopes,” Scumpia said.

Researchers developed researchers on a “convolutional neuro network” to transform RCM images of unstained hair into 3D images that look almost identical to the H&E (hematoxylin eosin), images used by dermatologists and dermatopathologists. Deep learning, a form of machine learning, constructs artificial neural networks that, like the human brain, can “learn” from large quantities of data.

“This framework is able to perform virtual histology for a variety of skin conditions, including basal cell carcinoma. It also provides detailed 3D images of various skin layers,” said Ozcan, who also holds UCLA faculty appointments in the fields of bioengineering and surgery. He also is the associate director of the California NanoSystems Institute. “Virtually stained images show the same color contrast and spatial features as traditional stained microscopic images taken from biopsied tissue” Ozcan added. This approach may allow medical professionals to see the total histological features of the skin without invasive skin biopsies or the time-consuming task of chemical processing and the labeling of tissue.”

Rubinstein claims that this is a proof-of-concept research. Dermatoscopes are the only tool that dermatologists currently use in clinics to assist them. They magnify the skin and make sunlight polarized. They’re not able to do much other than aid a dermatologist to see patterns,” said Rubinstein. He also employs reflectance confocal microscopes in clinic.

The authors stated that a number of steps remain in translating this technology into clinical applications however, their aim is to provide virtual histology technology that can be integrated into any device -; small, large, or integrated with other optical-imaging systems. Once the neural network is “trained,” with many tissue samples and the use of powerful graphics processing units (GPUs) and GPUs, it can run on a computer or network, enabling rapid transformation from a standard image into a virtual histology image.

Future studies will determine if this digital, biopsy-free approach can interface with whole-body imaging and electronic medical records to signal a new age of “digital dermatology” and alter the way dermatology is practiced. Furthermore, the team of researchers will examine whether this artificial intelligence platform can work with other AI technologies to identify patterns, and help in the clinical diagnosis.

Content Source:

Gemma Wilson

Gemma is a journalism graduate with keen interest in covering business news – specifically startups. She has as a keen eye for technologies and has predicted quite a few successful startups over the last couple of years.

Related Articles