Near-infrared (NIR) spectroscopy has become a popular technique for assessing food quality due to its advantages over complex chemical analysis methods. However, the application of NIR spectroscopy for evaluating fish quality based on urea content has not been extensively explored. This study investigates the use of NIR spectroscopy in combination with machine learning (ML) techniques to classify fish samples into two safety classes—Safe and Unsafe—based on their urea content. A comprehensive NIR dataset comprising 11,960 spectra collected from eight distinct positions within the fish body was obtained from 299 fish samples of mackerel, tuna, and pompano species. ML experiments were conducted to classify fish samples based on whether their urea content exceeded the permissible limit of 1000 ppm. To address class imbalance and optimize ML models, various data pre-processing and feature extraction techniques, as well as ML algorithms, were explored. The results demonstrated that utilizing NIR data specifically obtained from the outer skin of the stomach yielded superior models for fish safety classification. A feature extraction method employing pre-processed NIR spectra and their first derivatives, combined with an optimized convolutional neural network architecture, outperformed traditional ML classifiers, achieving an accuracy of up to 83.9%.

Spectral imaging has recently gained attention as a promising noninvasive technique for diagnosing human skin conditions, owing to its safety and noninvasive nature. However, one of the main challenges to advancing the application of this technique is the lack of a comprehensive and reliable database capturing the optical characteristics of different skin tones and conditions. Monte Carlo Simulation on Multi-layer Media (MCML) has proven effective in modeling light-tissue interactions and generating spectral reflectance data for biological tissues, particularly skin. However, the traditional simulation process is time-consuming. This study examines and validates the results of an accelerated MCML simulation using CUDA technology, by comparing its spectral diffuse reflectance outputs with actual measurements from healthy skin across three distinct skin tones within the 900 nm to 1700 nm wavelength range. The high degree of agreement between simulated and measured data supports the feasibility of building a robust MCML-based spectral reflectance database.