Body & Mind

AI-powered device to detect all 3 common skin cancers okayed for doctors

AI-powered device to detect all 3 common skin cancers okayed for doctors
The FDA has granted clearance to an AI-powered device that helps physicians detect skin cancers
The FDA has granted clearance to an AI-powered device that helps physicians detect skin cancers
View 1 Image
The FDA has granted clearance to an AI-powered device that helps physicians detect skin cancers
1/1
The FDA has granted clearance to an AI-powered device that helps physicians detect skin cancers

The FDA has granted clearance for the first AI-powered handheld medical device to assist physicians in detecting all three common skin cancers: basal cell carcinoma, squamous cell carcinoma, and melanoma. Providing a more accurate way of identifying skin cancer will enable patients to access necessary treatment more quickly.

According to current estimates, one in five Americans will develop skin cancer in their lifetime. The three most common skin cancers are basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and melanoma, the most dangerous form of skin cancer because of its tendency to spread.

If caught early, skin cancer is highly treatable, but it makes early diagnosis crucial. Until now, detection has relied on the naked eye or magnified visual examination of skin lesions, which is dependent upon clinical training and the subjective judgment of primary care physicians (PCPs). However, the FDA recently upped the detection game, granting clearance to DermaSensor, the first AI-powered device to detect all three common skin cancers in real time.

“We are entering the golden age of predictive and generative artificial intelligence in healthcare, and these capabilities are being paired with novel types of technology, like spectroscopy and genetic sequencing, to optimize disease detection and care,” said Cody Simmons, co-founder and CEO of DermaSensor. “Equipping PCPs, the most abundant clinicians in the county, to better evaluate the most common cancer in the country has been a major, long-standing unmet need in medicine. While dozens of companies have attempted to address this problem in recent decades, we are honored to be the first device cleared by the FDA that provides PCPs with an automated tool for evaluation of suspicious lesions.”

It's important to pause here to explain the difference between ‘FDA approved’ and ‘FDA cleared’. Medical devices intended for human use fall into three categories. Class III are complex, implanted devices and products, like pacemakers and breast implants. They require FDA approval, meaning their benefits outweigh the known risks for their intended use, proven using data from clinical trials. Lower-risk devices and products for external use fall into Class I or II (for example, the ECG app for the Apple Watch is in Class II). These devices are ‘cleared’ rather than approved, meaning the FDA has allowed a device to market because it’s “substantially equivalent” (in terms of safety and efficacy) to another legally marketed device.

The tip of the DermaSensor device uses elastic scattering spectroscopy (ESS), a process that evaluates how photons scatter when reflected off different cellular structures. Given malignant skin lesions’ different cellular and sub-cellular make-up, they’ve been reported to scatter light differently from benign lesions. DermaSensor’s built-in AI uses these spectroscopic images to provide physicians with information to help them assess skin lesions, a process explained in the video below.

How DermaSensor Works

The FDA’s clearance was based on clinical trials evaluating DermaSensor’s efficacy. In a study led by the Mayo Clinic across 22 centers in the US and Australia and published in the Journal of Clinical and Aesthetic Dermatology, researchers tested the device on 1,005 patients with, on average, one or two skin lesions. Before the lesions were biopsied, PCPs predicted which ones would be malignant. The DermaSensor device had an overall sensitivity of 95.5% for detecting malignancy – sensitivity of 87.5% for melanoma, 97.8% for BCCs and 98.7% for SCCs – compared to the PCPs’ overall sensitivity of 83.0%.

In an accompanying clinical validation study appearing in the same journal, 108 PCPs evaluated 50 skin lesions (25 malignant, 25 benign) with and without the device. The DermaSensor device was found to decrease the number of missed skin cancers from 18% to 9%. Additionally, physicians reporting high confidence in their evaluations increased from 73.0% to 81.6% when they used the device.

Better identifying skin cancer in the primary care setting will accelerate patient access to necessary treatment. In addition to improving primary care, given the device’s accuracy, DermaSensor is expected to improve PCPs’ collaboration with dermatologists by streamlining the referral system.

Source: DermaSensor

2 comments
2 comments
Smokey_Bear
I love reading things like this, where AI outperforms the best doctor diagnostic, because this is scalable...a great doctor is not.
Ric
All sounds great. One thing this article is missing is the percentage of false positives, a standard data point in these sorts of studies. Hopefully it’s because there were none ;)