Developing Low-Cost Mobile Device and Apps for Accurate Skin Spectrum Measurement via Low-Cost Spectrum Sensors and...

Area of Science:

• Biomedical Engineering
• Dermatology
• Computer Science (Artificial Intelligence)

Background:

• Skin spectral information is valuable in cosmetics and clinical medicine.
• Existing skin spectrum measurement devices are expensive and bulky, limiting practical applications.
• Previous work demonstrated convolutional neural network (CNN) effectiveness for Fitzpatrick skin-type classification.

Purpose of the Study:

• To develop a low-cost miniature skin spectrum measurement system (LMSSMS) using deep neural network (DNN) technology.
• To verify the feasibility and accuracy of the developed LMSSMS.
• To integrate multiple skin-spectrum-related applications into a mobile app.

Main Methods:

• Developed a miniature skin spectrum measurement device (MSSMD) using low-cost materials, 3D printing, and an LED light source.
• Designed a deep neural network (DNN) model to enhance measurement accuracy.
• Created a mobile application for system control, results display, and various skin analyses.

Main Results:

• The developed LMSSMS achieved an accuracy of 94.7% when compared to standard instrumentation.
• The system successfully integrates functionalities like erythema/melanin index measurement, skin-type classification, and sun-exposure estimation.
• The mobile app provides a user-friendly interface for controlling the device and viewing results.

Conclusions:

• The developed low-cost miniature skin spectrum measurement system (LMSSMS) is feasible and accurate.
• This technology has significant potential for wider application in cosmetics, clinical medicine, and personal health monitoring.
• The integration of DNN and mobile technology offers a promising direction for accessible skin analysis.