Machine Learning-Driven Thermal Imaging for Classifying Skin Response to Infrared Laser Irradiation

This study investigates the thermal effects of infrared laser irradiation on skin samples using advanced infrared thermography and machine learning techniques. The experimental setup involved exposing chicken skin samples to laser wavelengths of 808 nm, 830 nm, and 980 nm, while capturing thermal images to analyze temperature variations. A total of 3,792 thermal images were collected across four experimental categories, including a control group with no laser exposure. The data were processed using a Convolutional Neural Network (CNN) implemented in Python, which achieved training and validation accuracies of 94% and 96%, respectively. The results indicated that laser exposure led to temperature increases ranging from 1 to 6 °C, depending on the wavelength used. This research demonstrates the potential of combining thermal imaging with machine learning to enhance the understanding of skin's thermal response to laser treatments, providing valuable insights for optimizing clinical applications and improving patient safety in laser-based therapies.