Yang Albert F, Patel Soham, Chun Keum San, Richards Dylan, Walter Jessica R, Okamoto Kazuaki, Paller Amy S, Ikoma Akihiko, Xu Shuai
Department of Dermatology, University of Michigan, Ann Arbor.
University of Kansas School of Medicine, Kansas City.
JAMA Dermatol. 2025 Apr 1;161(4):406-410. doi: 10.1001/jamadermatol.2024.5697.
Although more than 1 in 10 people experience pruritus, there are limited medical technologies that can accurately and continuously quantify and simultaneously reduce scratching behaviors through nonpharmacological methods.
To evaluate the accuracy and efficacy of an artificial intelligence-enabled wearable sensor with closed-loop haptic feedback to decrease nocturnal scratch in patients with mild atopic dermatitis who report a moderate to severe degree of scratching.
DESIGN, SETTING, AND PARTICIPANTS: This single-arm 2-stage cohort study with a within-participants design was conducted at a single center and carried out in an at-home environment. Adult patients with atopic dermatitis were recruited from the Northwestern University Department of Dermatology in Chicago, Illinois. Participants were fluent in English, 18 years old or older, had a diagnosis of atopic dermatitis, and self-reported moderate or severe scratching behaviors. Each participant's disease at time of recruitment was scored via the Validated Investigator Global Assessment for Atopic Dermatitis. Data were collected from April to July 2023.
Haptic feedback delivered by a wearable sensor mounted on the hand triggered whenever nocturnal scratch was detected by an artificial intelligence algorithm. Participants initially wore the sensor for sensing only for 7 nights to assess baseline nocturnal scratching and sleep parameters. This was followed by an additional 7 nights of wearing the sensor with haptic feedback activated.
Retrospective analysis was performed for scratch events and scratch duration per night and per hour of sleep opportunity. Paired t tests were used to compare changes in patient scratching behaviors before and after use of the artificial intelligence-enabled haptic feedback devices.
Of 10 included patients, 6 were female, and the mean (SD) age was 36 (12) years. All patients had a Validated Investigator Global Assessment for Atopic Dermatitis score of 0 to 2 (clear to mild) who contributed a total of 104 sleep nights and 831 monitoring hours. No patients were lost to follow-up. There was a significant decrease in mean (SD) scratch events nightly (45.6 [24.0] vs 32.8 [13.0]; P = .03), a 28% difference, and mean (SD) scratch duration per hour of sleep opportunity (15.8 [10.7] seconds vs 7.9 [3.7] seconds; P = .01), a 50% difference, when haptic feedback was activated in the second week without a decrease in total sleep opportunity.
This study found that haptic feedback may be used as a nonpharmacological intervention to reduce nocturnal scratching in patients with mild atopic dermatitis. Future randomized studies are needed to confirm.
尽管超过十分之一的人会经历瘙痒，但通过非药物方法能够准确且持续地量化并同时减少抓挠行为的医学技术有限。
评估一种具有闭环触觉反馈的人工智能可穿戴传感器在减少报告有中度至重度抓挠程度的轻度特应性皮炎患者夜间抓挠方面的准确性和有效性。
设计、地点和参与者：这项采用参与者内设计的单臂两阶段队列研究在单一中心进行，并在家庭环境中开展。成年特应性皮炎患者从伊利诺伊州芝加哥西北大学皮肤科招募。参与者英语流利，年龄在18岁及以上，被诊断为特应性皮炎，且自我报告有中度或重度抓挠行为。每位参与者在招募时的疾病通过特应性皮炎经验证的研究者整体评估进行评分。数据于2023年4月至7月收集。
每当人工智能算法检测到夜间抓挠时，安装在手上的可穿戴传感器就会提供触觉反馈。参与者最初仅佩戴传感器进行7晚的感知，以评估基线夜间抓挠和睡眠参数。随后再佩戴传感器7晚，同时激活触觉反馈。
对每晚以及每小时睡眠机会中的抓挠事件和抓挠持续时间进行回顾性分析。采用配对t检验比较使用人工智能触觉反馈设备前后患者抓挠行为的变化。
纳入的10名患者中，6名女性，平均（标准差）年龄为36（12）岁。所有患者的特应性皮炎经验证的研究者整体评估评分为0至2（清除至轻度），共贡献了104个睡眠夜晚和831个监测小时。无患者失访。在第二周激活触觉反馈时，每晚平均（标准差）抓挠事件显著减少（45.6 [24.0] 次对32.8 [13.0] 次；P = 0.03），差异为28%，每小时睡眠机会的平均（标准差）抓挠持续时间也显著减少（15.8 [10.7] 秒对7.9 [3.7] 秒；P = 0.01），差异为50%，且总睡眠机会未减少。
本研究发现触觉反馈可作为一种非药物干预措施来减少轻度特应性皮炎患者的夜间抓挠。未来需要进行随机研究以证实。
