A wearable sensor that sticks onto the back of the hand and uses an artificial intelligence (AI) algorithm to monitor scratching activity when the wearer sleeps has been shown to reduce subconscious nighttime scratching in patients with atopic dermatitis (AD) comparable with the effect of topical medications, researchers have reported.
A small cohort study of 10 adults with mild AD who wore the device at night reported a 28% reduction in scratch events nightly, from a mean of 45.6 to 32.8 (P = .03), and a 50% reduction in scratch duration, from a mean of 15.8 to 7.9 seconds (P = .01). The findings were published online on February 5 in JAMA Dermatology.
“The results were more substantial than we anticipated,” lead study author Shuai (Steve) Xu, MD, professor of dermatology at Northwestern University, Chicago, and a patent holder on the device, told Medscape Medical News.
“We saw a 40% drop in the total scratch duration per night and a 50% drop in scratch duration per hour of sleep opportunity,” said Xu, who is also an engineer. “That’s a pretty big drop when you think about it. It’s on par with certain powerful medicines.”
How the Sensor Works
The sensor is about the size of a quarter, Xu said, and it sticks to the back of the hand like an adhesive bandage. It is embedded with an AI algorithm that has been trained on more than 10,000 different scratch events, “so it’s 99% accurate,” he added. When it senses a scratch — when the fingernails on the hand touch skin — it delivers a gentle vibration to the hand, but one that’s not strong enough to wake up the wearer, he said. The sensor also demonstrated a sensitivity of 95% and a specificity of 99%.
In the study, participants, whose mean age was 36 years, initially wore the sensor for sensing only for seven nights to gather baseline information on nighttime scratching and sleep patterns. Then they wore the device for another seven nights, this time with the haptic feedback activated when it sensed a scratch. All study participants had a Validated Investigator Global Assessment for Atopic Dermatitis score of 0-2 (clear to mild) and reported they had moderate or severe scratching episodes.
During the first week, when the sensor was used without haptic feedback, each participant-night generated a mean total scratch duration of 96.5 seconds, 45.6 total scratch events, and 2 seconds of scratch duration per scratch event. In the second week, when haptic feedback was activated, those respective measures were a mean of 58.1 seconds, 32.8 total scratch events, and 1.75 seconds per scratch event.
When normalized by hour, haptic feedback resulted in a 39% decrease in mean total scratch events per sleep opportunity hour (from 7.4 to 4.5, P = .004) and a 50% reduction in mean total scratch duration per sleep opportunity hour (from 15.8 to 7.9 seconds, P = .01).
The study authors acknowledged that the small sample size was a limitation, and the presence of the sensor on the back of the hand itself may have contributed to reducing scratching behavior with or without haptic feedback.
Another limitation is that the sensor was placed only on the dominant hand, and it only detects activity in the hand on which it’s worn, Xu told Medscape Medical News. “It’s an interesting hypothesis,” he said. “We are thinking that your brain might get trained when it doesn’t know which hand is going to scratch, but it doesn’t want that slight buzz.” He said further study should provide that answer.
Because the sensor uses AI, it doesn’t require any training or calibrating for wearers. “It works out of the box,” he said. “That’s enabled by AI.”
And while AI algorithms typically require expensive data chips, Xu said his team was able to “make it really efficient to put it right on the sensor. It’s an advance in engineering that makes the AI algorithm very light, and it doesn’t draw a lot of power.”
‘Additive Benefit’ With Medications
At this point, Xu said he doesn’t anticipate such a device will supplant topical medications altogether in treating AD. “It has the potential to be an additive benefit,” he said. “This can be something on top of a medication to prevent excessive scratching subconsciously.”
The next step in the group’s research is to conduct a larger randomized control trial, Xu said.
“This study opens the field for use of biofeedback approaches at home to reduce itch and stress and improve sleep in numerous types of itchy conditions,” Gil Yosipovitch, MD, professor of dermatology and director of the Miami Itch Center at the University of Miami, Coral Gables, Florida, told Medscape Medical News. “Biofeedback can help patients to relax and may improve AD that flares with stress.”
Having a nonmedical alternative for mild AD would be significant, Yosipovitch added. “There are patients with AD who have itch associated with neural sensitization and minimal rash, and therefore having a nonpharmacological approach to reduce scratching would be of benefit,” he said.
Future studies should be of a larger scale and assess of autonomic nerve function, such as sympathetic/parasympathetic, using heart rate variability, he added.
Having a wearable as an alternative to topical medications to treat AD is “very important,” said Paras Vakharia, MD, PharmD, assistant professor of dermatology at Northwestern University, who was not affiliated with the study.
“There are numerous things to consider with topical medications,” Vakharia told Medscape Medical News. “These include side effects even from topical application, recurring costs of a monthly prescription, the burden of having to apply something to the skin one to two times a day regularly; and there are also patients for whom topicals may be ineffective, and we are going back and forth spending a lot of time trying to find topicals that work.
“So, to have an alternative, particularly a non-pharmacologic alternative, is very beneficial.”
Maruho Co. provided funding for the study. Xu is the founder and CEO of Sibel Health, which is developing the wearable, and holds a patent on it. Vakharia and Yosipovitch had no relevant financial relationships to disclose.
Richard Mark Kirkner is a medical journalist based in Philadelphia.
Source link : https://www.medscape.com/viewarticle/stick-ai-hand-sensor-reduces-nighttime-scratching-mild-2025a10002wl?src=rss
Author :
Publish date : 2025-02-06 03:24:36
Copyright for syndicated content belongs to the linked Source.