NIR spectroscopy combined with machine learning as a time-tested tool of analytical chemistry

In the area of vibrational spectroscopy augmented by Machine Learning (ML) and Artificial Intelligence (AI) methods, Near-Infrared (NIR) spectroscopy takes its well-deserved place at the forefront. Since its beginnings, the profound potential of NIR spectroscopy in both quantitative and qualitative analysis, was tightly coupled with the evolution of ML methods. A considerable number of ML methods have been specifically developed for - and in conjunction with - NIR spectroscopy, to meet the needs arising from its applications [1]. Today, novel instrumentation (e.g., miniaturized spectrometers, UAV-based sensors) and a new landscape of applications (from industrial online applications, through medicinal plants and crops monitoring to everyday consumer use) create high demand and depend on the new, potent, robust, and user-friendly data analytical and AI tools A compelling example of this evolution can be found in the advancement of instrumentation: the emergence of highly-miniaturized and cost-effective NIR spectrometers in the form of pocket-sized sensors. When combined with modern data analytical frameworks, including cloud-based spectral analysis and pre-calibrated ML models, these pocket-sized spectrometers enable everyday consumers to acquire information on the quality, authenticity of food products or medicines, without any need for professional training. This presentation offers a comprehensive exploration of the current analytical potential of NIR spectroscopy combined with AI, illustrating its key applications in the fields of food quality assessment, agriculture, natural medicines, and environmental monitoring. This presentation will cover the current immense analytical potential of the NIR spectroscopy combined with AI methods, with examples of key applications in food, agriculture, natural medicines, and environmental analysis [3]. [1] Near-infrared spectroscopy, Ozaki, Y., Huck, C.W., Tsuchikawa, S., Engelsen, S.B., Eds.; Springer: Singapore, 2021