Summary: UCLA researchers in the Department of Electrical and Computer Engineering have developed a flexible, cost-effective, AI-enabled wearable sensor that facilitates early, non-invasive diagnosis of allergic contact dermatitis.
Background: Allergic contact dermatitis (ACD) is a hypersensitivity reaction of the skin triggered by direct contact with specific allergens. The condition affects 20% of the population yet is diagnosed ineffectively and rudimentarily. The state-of-the-art diagnostic method for ACD is patch testing, which involves applying standardized allergen panels to the skin followed by clinical evaluation. While patch testing remains the clinical gold standard, it is inherently invasive as it deliberately provokes localized dermatitis to identify contact allergens via inflammatory response. The method is further limited by extended diagnostic timelines, requirement for multiple clinical visits over several days, and patient discomfort due to induced skin irritation. To address these limitations, a non-invasive, convenient, and real-time diagnostic device is urgently needed to accurately identify ACD without relying on prolonged inflammation or extended clinical monitoring.
Innovation: Professor Ozcan and his research team have innovated a wearable and cost-effective sensor system capable of measuring skin optical properties, enabling early, non-invasive detection of allergic contact dermatitis (ACD). The device can identify early inflammatory changes in the skin before visible symptoms fully manifest, providing improved diagnostic capabilities for clinicians. The sensor is designed with high signal-to-noise performance, integrated into a low-power, flexible, and water-resistant device suitable for continuous wear in real-world conditions. Data acquisition occurs wirelessly in a time-lapsed format, enabling continuous monitoring without disrupting patient activity. The collected data is processed through a deep neural network (DNN) to classify skin responses in real time. Positive reactions are immediately communicated to the user via LED indicators, and the DNN can further interface with a sequential forward feature selection algorithm to optimize data collection, minimizing measurement redundancy while maintaining diagnostic accuracy. This technology has the potential to transform ACD diagnosis from a reactive, clinic-based procedure to a proactive, point-of-care capability, improving patient comfort, accessibility, and treatment outcomes. Beyond its primary application, the technology holds potential for adaptation across numerous disease states and diagnoses, enabling it to function as a broadly applicable clinical platform.
Potential Applications: ● Diagnosis and monitoring of ACD ● Inflammatory skin condition screening ● Occupational health surveillance for allergen exposure ● Research tool for studying allergen responses ● Remote patient monitoring in teledermatology ● Consumer skincare and cosmetic testing for adverse reactions ● Developing nation disease monitoring
Advantages: ● Diverse skin type applications ● Cost-effective and scalable ● Flexible, lightweight, and comfortable for extended wear ● Low-power for long-term use ● Non-invasive, avoiding skin provocation ● Real-time wireless monitoring with AI-driven analysis
State of Development: The wearable sensors have been successfully validated on diverse skin phantoms, demonstrating reliable performance.
Reference: UCLA Case No. 2025-301
Lead Inventor
Aydogan Ozcan, Professor, Department of Electrical & Computer Engineering