Getting screened for skin cancer isn’t easy. You’re asked to strip completely and don a hospital gown before the dermatologist spends ten minutes examining you from stem to stern. Skin cancer is crafty at concealment. That means the doctor must look at your scalp, your soles, your genital area, and between your buttocks. Suspicious spots, moles, and other potential problems are targeted for biopsy. Scraped off or sliced away, the tissue sample is sent to a lab while you anxiously wait days for the results.
Patients–and doctors–have long wondered if there was a better way. Engineers and medical researchers at the University of California, Los Angeles agreed. Now they are developing new imaging technology that may reduce the need for skin biopsies. Here’s how it works.
Skin cancer and screenings
As with all cancers, skin cancer occurs as the result of ongoing genetic mutations. Although most of the time damaged DNA is not passed on, with cancer these cells continue to divide. Left unchecked, cancer cells can spread and even attack healthy cells. It’s important to note, however, that not all cancers spread and not all cancerous tumors are life threatening.
Although skin cancer isn’t the most fatal version of the disease, it is the most common. Around one out of five people will be diagnosed with skin cancer sometime in their life. Every year in the United States alone, almost 200,000 people are diagnosed with melanoma––the most deadly skin cancer. However, even this dangerous form of the disease has a 99% five-year survival rate––assuming it doesn’t spread to the lymph nodes.
Although U.S. melanoma rates doubled from 1982 to 2011, its incidence has begun declining with people under age 30. While the number of deaths climbed during that period, from 2013 to 2016, the number of melanoma deaths among whites declined by nearly 20% with the sharpest drop among those over age 50. Screening and early detection gets some credit for the reduction in death rates, but innovative treatments are primarily responsible. Still, for anyone with fair skin, red hair, family history of skin cancer, or childhood sunburns, regular screenings are recommended. Unfortunately, not only do many people skip screenings, research during the COVID-19 pandemic shows a huge drop in patients coming in for examinations. At UCLA they may be coming up with an easier way to do it.
Pain-Free Screening Tool
A collaboration between UCLA Samueli School of Engineering and the David Geffen School of Medicine may lead to a new way of diagnosing skin cancer. Researchers deployed a subset of machine learning––where a computer system learns and adapts without further information from the programmer. Instead, it relies on algorithms and statistical models. The subset they deployed is called a convolutional neural network. Just like the neural network in the body, the one in this computer system is interconnected––with each one having an input and output node. When an individual node’s output exceeds a certain value, it’s activated and passes along the data to the next node. If it doesn’t reach the threshold, the data isn’t delivered.
For the last three years, the team at UCLA has used images of intact skin that were taken by a noninvasive optical technology. This took black and white images––the lack of color would be challenging even for skilled diagnosticians. That’s why the tool transformed them into 3D images that would be more familiar to dermatologists and dermatopathologists. It’s called “virtual histology” or the study of the microscopic structure of tissues. Utilizing this novel image tech means bypassing “… several standard steps typically used for diagnosis—including skin biopsy, tissue fixation, processing, sectioning and histochemical staining. Images appear like biopsied, histochemically stained skin sections images on microscope slides,” explains the study’s senior author, Aydogan Ozcan, Chancellor’s Professor and Volgenau Chair for Engineering Innovation of the Electrical and Computer Engineering Department at UCLA Samueli. Dr. Philip Scumpia, assistant professor of dermatology and dermatopathology at the David Geffen School of Medicine at UCLA, adds, “The current standard for diagnosing skin diseases, including skin cancer, relies on invasive biopsy and histopathological evaluation. For patients, this often leads to unnecessary biopsies and scars as well as multiple visits to doctors. It also can be costly for patients and the health care system.”
Although it’s still in clinical trials, the images resemble biopsied skin placed under a microscope. The benefits are more than just reduced cost and discomfort for patients. The less unpleasant a screening process, the more people will participate––which over time could increase the number of cancers caught early enough to treat. That raises their survivability. At the moment, however, there’s no available tech that will let you keep your clothes on.
Written by John Bankston
References
- Skin cancer screening: What to expect
- DNA damage and gene mutations
- Neoplasm (Tumor)
- Skin cancer
- US Deaths from Melanoma Drop Substantially – National Cancer Institute
- Cancer screenings decline significantly during pandemic
- Biopsy-free in vivo virtual histology of skin using deep learning | Light: Science & Applications
- What are Convolutional Neural Networks? | IBM
- New imaging technology may reduce need for skin biopsies
