AI for Skin Cancer Diagnosis
DermaSensor from DermaSensor Inc.
FDA Clears Device with BU-Developed Technology That Makes Skin Cancer Detection Easier
DermaSensor uses optical technique pioneered by BU biomedical engineer Irving J. Bigio and could cut number of missed cancers by half
Maybe it’s just a funky-looking, unique-to-you mole. But that irregular patch or evolving mark could signal bad news: skin cancer, the most common form of cancer in the United States. Although spotting skin cancer early could save your life, it can be tough for even some medical professionals to judge if a mark is benign or potentially harmful. A new noninvasive skin cancer detection device—powered by technology pioneered by a professor at Boston University’s College of Engineering—aims to make telling the difference easier and faster.
The US Food & Drug Administration recently cleared for US markets DermaSensor, which uses light and artificial intelligence to examine skin lesions and assess whether a patient should be referred to a specialist. The company says it has the potential to slash the number of missed skin cancers by half. DermaSensor’s underlying sensing technology, elastic scattering spectroscopy (ESS), was developed and refined by Irving J. Bigio, an ENG professor of biomedical engineering and of electrical and computer engineering. He’s a scientific advisor to the eponymous company behind the device, which also licensed patents from Bigio and BU.
“The FDA had designated this as a breakthrough technology, which means they gave it higher priority for review because they see it as having a real impact,” says Bigio. “And the trials showed that it actually does work.”
According to the American Academy of Dermatology Association, one in five of us will grapple with skin cancer at some point in our lives, which is why it recommends regular skin exams. In its pivotal FDA study—the research that makes or breaks a new clinical technology—DermaSensor says researchers found the device had “a sensitivity of 96 percent across all 224 skin cancers.” It can detect the most frequent forms of skin cancer—basal cell carcinoma and squamous cell carcinoma—and the less common, but more deadly, melanoma.
In the clinic, a physician or nurse puts the tip of the DermaSensor on a lesion. The device then fires off a pulse of light and analyzes the spectral information of the backscattered light using an AI-powered algorithm. Eladio Rodriguez-Diaz (ENG’09), a former PhD student in Bigio’s lab, developed much of the sensor’s machine learning and data analysis technology; he’s a coinventor on some of the patents.
“It’s incredibly gratifying to see Dr. Bigio’s innovative research incorporated into an FDA-cleared medical device, especially one with the potential to noninvasively detect skin cancer,” says Frances Forrester, director of business development in BU Technology Development. “Early detection is known to save lives, and now a new tool is available to US-based primary care providers and their patients through BU research.”
Bigio says the device is the first consumer-facing medical product using ESS to hit the market, but that he and his clinical collaborators have spent decades testing the technology’s potential in other fields. In multiple National Institutes of Health–funded studies, they’ve shown it could help pinpoint the locations of tumors, measure the effectiveness of cancer medications, detect malignant thyroid nodules, and differentiate normal from abnormal polyps during a colonoscopy. In some cases, he says, that research is ready to make the jump from the bench to the bedside—they just need to find the right commercial partners to take things to the next level, as DermaSensor has done.
“I think the DermaSensor success in getting FDA clearance, and some initial commercial success that I’m quite confident is going to come now, will be the rising tide that floats other boats,” says Bigio. “Once investors or med tech companies see this, they’re going to take a stronger interest. And we’re already starting to see that in what we’ve been doing for interventional radiology and intravital measurements in various organs.”
Sources from the following weblinks:
DermaSensor, Inc
https://www.dermasensor.com/
SkinVision
SkinVision is an AI-powered application dedicated to helping users assess their skin spots for potential signs of skin cancer. It swiftly provides a risk assessment, usually within 30 seconds, by analyzing photos of skin spots uploaded by users. The app’s intelligent algorithm scrutinizes these images for indicators of skin cancer, such as irregular patterns, multiple colors, asymmetry, size, and uneven borders. It’s essential to note that SkinVision serves as an early warning system and is not a substitute for professional medical advice.
Features and benefits
AI-powered analysis: Utilizes advanced AI technology to meticulously analyze skin spots for potential signs of skin cancer, delivering highly accurate results.
Risk assessment: Swiftly provides risk assessments for skin spots, categorizing them as low risk, low risk with symptoms, or high risk within a mere 30 seconds.
Healthcare efficiency: Aims to reduce unnecessary doctor visits by accurately identifying spots that may not require immediate medical attention, potentially saving on healthcare costs.
User-friendly: Designed with simplicity in mind, allowing users to effortlessly capture photos of their skin spots and receive immediate feedback.
AI Diagnosis Algorithm for Skin Cancer Developed at Stanford
Deep learning algorithm does as well as dermatologists in identifying skin cancer
January 25, 2017
In hopes of creating better access to medical care, Stanford researchers have trained an algorithm to diagnose skin cancer. They made a database of nearly 130,000 skin disease images and trained their algorithm to visually diagnose potential cancer. From the very first test, it performed with inspiring accuracy.
https://news.stanford.edu/2017/01/25/artificial-intelligence-used-identify-skin-cancer/
Comments: