AI in Optical Spectroscopy Analysis

Topic Information

Dear Colleagues,

Spectroscopic techniques are widely employed in the field of analytical chemistry. Common modalities, including ultraviolet/visible/near-infrared/mid-infrared spectroscopy, Raman spectroscopy, laser-induced breakdown spectroscopy, terahertz spectroscopy, fluorescence spectroscopy, and their corresponding imaging modes, have been extensively investigated for chemical analysis across diverse application domains such as food, agriculture, pharmacy, and medicine. These techniques operate on distinct principles and working mechanisms. Notably, compared to traditional wet chemistry methods, spectroscopic techniques offer advantages such as rapid analysis, non-destructive or minimally invasive measurement, low cost, and environmental friendliness.

Based on specific instruments, these spectroscopic techniques can generate large volumes of data. The methods for interpreting this data vary. Direct analysis of certain spectral features can provide intuitive information about the sample. However, for more complex qualitative analyses (e.g., classification) and quantitative analyses (e.g., determining component concentrations), appropriate spectral data analysis methods are essential for processing the acquired data. The required processing techniques also differ depending on the specific type of spectral data. Chemometric methods have achieved significant success in the field of spectral data analysis.

The development of artificial intelligence (AI), particularly machine learning and deep learning, has provided a continuous stream of inspiration and solutions for spectral data analysis. The application of AI in spectroscopic analysis primarily addresses three major challenges inherent in traditional methods: cumbersome data processing, difficulty in feature extraction, and limited model generalization capability. Consequently, AI enhances the accuracy and robustness of spectroscopic detection.

Despite their clear advantages, spectroscopic techniques are susceptible to variations in equipment (e.g., parameters, models), environmental conditions, and sample characteristics. These factors have largely confined their use to research settings, leaving a gap towards practical deployment and application. Furthermore, for quantitative analysis of components, obtaining reference values entails substantial cost. Optimizing and improving methodologies to reduce the waste associated with repetitive measurements are also necessary. Therefore, the primary objective and direction of current research is to leverage the strengths of AI to overcome the challenges hindering the practical application of spectroscopic techniques.

The primary objective of this Topic is to address the challenges in applying spectroscopic techniques by integrating artificial intelligence algorithms, tailored to the characteristics of different spectroscopic modalities. Building upon the substantial body of existing research, the aim is to progressively advance these techniques from feasibility studies and laboratory research towards practical deployment and application.

Dr. Chu Zhang
Dr. Jiyu Peng
Topic Editors

Keywords

Participating Journals

Journal Name	Impact Factor	CiteScore	Launched Year	First Decision (median)	APC
Agriculture agriculture	4.5	7.8	2011	17.4 Days	CHF 2600	Submit
AI Chemistry aichem	-	-	2026	15.0 days *	CHF 1000	Submit
Analytica analytica	7.4	7.2	2020	19.4 Days	CHF 1200	Submit
Applied Sciences applsci	2.9	6.1	2011	15 Days	CHF 2400	Submit
Chemosensors chemosensors	4.4	8.1	2013	19.8 Days	CHF 2000	Submit
Foods foods	6.0	10.3	2012	14.8 Days	CHF 2900	Submit
Plants plants	4.7	8.5	2012	14.8 Days	CHF 2700	Submit
Sensors sensors	4.0	9.4	2001	17.8 Days	CHF 2600	Submit

* Median value for all MDPI journals in the first half of 2026.

Preprints.org is a multidisciplinary platform offering a preprint service designed to facilitate the early sharing of your research. It supports and empowers your research journey from the very beginning.

MDPI Topics is collaborating with Preprints.org and has established a direct connection between MDPI journals and the platform. Authors are encouraged to take advantage of this opportunity by posting their preprints at Preprints.org prior to publication:

1. Share your research immediately: disseminate your ideas prior to publication and establish priority for your work.
2. Safeguard your intellectual contribution: Protect your ideas with a time-stamped preprint that serves as proof of your research timeline.
3. Boost visibility and impact: Increase the reach and influence of your research by making it accessible to a global audience.
4. Gain early feedback: Receive valuable input and insights from peers before submitting to a journal.
5. Ensure broad indexing: Web of Science (Preprint Citation Index), Google Scholar, Crossref, SHARE, PrePubMed, Scilit and Europe PMC.

Published Papers (5 papers)

Order results

Content type Publication Date

Result details

Normal Extended Compact

Journals

All Agriculture AI Chemistry Applied Sciences Foods Sensors Plants Chemosensors Analytica

Show export options expand_more Show export options expand_less

Select all

Export citation of selected articles as:

Plain Text BibTeX BibTeX (without abstracts) Endnote Endnote (without abstracts) Tab-delimited RIS

Error

Oops... you haven't selected anything for export.

Ok

clear

attachment

Supplementary material:
Supplementary File 1 (ZIP, 127 KB)

40 pages, 1167 KB  

Open AccessReview

A Preliminary Food–AI Readiness Framework for AI-Enabled Hyperspectral Systems in Fungal and Mycotoxin Risk Management: A Critical Narrative Review

by Ömer Faruk Yeşil

Foods 2026, 15(16), 2860; https://doi.org/10.3390/foods15162860 - 17 Aug 2026

Viewed by 215

Abstract

Fungal contamination and mycotoxins remain important hazards in cereal, nut, oilseed, and processed-food chains. For grain handlers, processors, storage facilities, quality-assurance laboratories, and technology developers, the practical challenge is not only detection but determining when an AI-enabled optical output is sufficiently documented to [...] Read more.

Fungal contamination and mycotoxins remain important hazards in cereal, nut, oilseed, and processed-food chains. For grain handlers, processors, storage facilities, quality-assurance laboratories, and technology developers, the practical challenge is not only detection but determining when an AI-enabled optical output is sufficiently documented to support accepting, holding, sorting, or referring material for confirmatory analysis. This critical narrative review, supported by structured evidence appraisal, examines artificial intelligence (AI)-enabled hyperspectral imaging (HSI) for fungal and mycotoxin risk management. The focused evidence map comprises a combined 20-study evidence map, including 17 spectral-imaging studies and three contextual non-imaging near-infrared studies addressing aflatoxins, fumonisins, deoxynivalenol (DON), zearalenone (ZEN), ochratoxin A (OTA), fungal status, and co-contamination; no patulin-specific AI-HSI study met the defined corpus criteria. Most evidence remains laboratory-based, with controlled contamination, limited independent validation, incomplete transfer testing, or no line-level demonstration. The review proposes the preliminary Food–AI Deployment-Readiness Framework (F-AIDRF), a non-regulatory appraisal tool covering analytical validity, algorithmic robustness, explainability and auditability, transferability and maintenance, operational feasibility, and decision integration. Applied descriptively, F-AIDRF is intended to reveal missing evidence, not to certify technologies. For food-industry readers, it offers a practical vocabulary for comparing screen-sort-confirm systems before investment, pilot testing, or regulatory-facing use. Full article

(This article belongs to the Topic AI in Optical Spectroscopy Analysis)

►▼ Show Figures

Figure 1

21 pages, 4653 KB  

Open AccessArticle

Soil Organic Carbon Estimation Using Dual-Interval Synergistic Selection and Overlap-Constrained Ridge Regression

by Anan Tao, Yuxi Ma, Chaoxu Yu, Jie Wang, Liuye Cao, Wenwen Kong and Fei Liu

Agriculture 2026, 16(16), 1700; https://doi.org/10.3390/agriculture16161700 - 8 Aug 2026

Viewed by 256

Abstract

Soil organic carbon (SOC) is a key indicator of soil quality, farmland productivity, and the terrestrial carbon cycle. Visible and near-infrared (Vis-NIR) spectroscopy offers a rapid approach for SOC estimation, but wavelength point selection may disrupt continuous spectral structures, whereas conventional wavelength interval [...] Read more.

Soil organic carbon (SOC) is a key indicator of soil quality, farmland productivity, and the terrestrial carbon cycle. Visible and near-infrared (Vis-NIR) spectroscopy offers a rapid approach for SOC estimation, but wavelength point selection may disrupt continuous spectral structures, whereas conventional wavelength interval selection may fail to fully exploit complementary information across spectral regions. In this study, a synergistic interval-constrained Ridge regression framework, termed sicRidge, was developed for SOC prediction. Continuous candidate intervals were generated using a sliding-window strategy, and a dual-interval synergistic search with an overlap constraint was applied to identify complementary and low-redundancy interval combinations. The selected intervals were then used to construct Ridge regression models. Using Vis-NIR spectra from 168 soil samples, sicRidge was compared with full-spectrum Ridge regression, five wavelength point selection-based Ridge models, and several wavelength interval selection-related benchmark models. sicRidge achieved the best prediction performance using 140 selected bands, with an R²_P of 0.834, RMSE_P of 2.010 g kg⁻¹, RPD of 2.483, and RPIQ of 3.777. The optimal intervals were 570~649 nm and 1880~1939 nm. These results indicate that sicRidge can improve SOC prediction by preserving continuous spectral structures while exploiting complementary cross-region information. Full article

(This article belongs to the Topic AI in Optical Spectroscopy Analysis)

►▼ Show Figures

Figure 1

17 pages, 487 KB  

Open AccessArticle

Near-Infrared Spectroscopy Non-Destructive Detection Modeling for Starch Content in Kernels of 58 Rainfed Corn Varieties

by Xiaoguang Yan, Guoliang Wang, Zhiyuan Ma, Liting Qi and Yanwei Du

Foods 2026, 15(15), 2599; https://doi.org/10.3390/foods15152599 - 24 Jul 2026

Viewed by 308

Abstract

Traditional methods for determining starch content in corn kernels are labor-intensive, destructive, and inefficient. To overcome these challenges, this work developed a rapid, non-destructive approach based on near-infrared hyperspectral imaging, applied to 58 rainfed corn varieties. A spectral preprocessing scheme combining wavelet transform, [...] Read more.

Traditional methods for determining starch content in corn kernels are labor-intensive, destructive, and inefficient. To overcome these challenges, this work developed a rapid, non-destructive approach based on near-infrared hyperspectral imaging, applied to 58 rainfed corn varieties. A spectral preprocessing scheme combining wavelet transform, multiplicative scatter correction, and standard normal variate transformation was employed to enhance spectral quality. A two-stage wavelength selection framework was established using competitive adaptive reweighted sampling and sparrow search algorithm optimization. From the selected optimal wavelengths, four predictive models, namely partial least squares regression, artificial neural network (ANN), convolutional neural networks, and gradient boosting decision tree, were established, implemented, and systematically compared. The results identify 14 key wavelengths (1020.65–1647.71 nm) strongly correlated with starch content, with clear assignments to specific chemical bonds and good physical interpretability. Among these models, the ANN exhibited the best performance. The R², RMSE, and RPD of the test set were 0.826, 0.759%, and 2.40, respectively, indicating favorable prediction accuracy and generalization ability. These key wavelengths provide a foundation for developing portable detection instruments. This work supports corn quality grading, breeding of high-starch varieties, and rapid raw material screening, thereby enhancing the quality and efficiency of the corn industry. Full article

(This article belongs to the Topic AI in Optical Spectroscopy Analysis)

►▼ Show Figures

Figure 1

attachment

Supplementary material:
Supplementary File 1 (ZIP, 54 KB)

33 pages, 7765 KB  

Open AccessArticle

UAV Multispectral Estimation of Citrus Leaf Nitrogen Content by Integrating Object-Based Canopy Extraction and PSO-Optimized Machine Learning

by Hongmei Gu, Weiqi Zhang, Yuliang Fu, Yun Zhong and Songlin Wang

Agriculture 2026, 16(15), 1570; https://doi.org/10.3390/agriculture16151570 - 23 Jul 2026

Viewed by 487

Abstract

Leaf nitrogen content (LNC) is an important physiological indicator for evaluating citrus nutritional status, photosynthetic capacity, and fertilization demand. However, conventional LNC determination mainly relies on field sampling and laboratory chemical analysis, which are destructive, time-consuming, labor-intensive, and limited in spatial continuity, making [...] Read more.

Leaf nitrogen content (LNC) is an important physiological indicator for evaluating citrus nutritional status, photosynthetic capacity, and fertilization demand. However, conventional LNC determination mainly relies on field sampling and laboratory chemical analysis, which are destructive, time-consuming, labor-intensive, and limited in spatial continuity, making them unsuitable for large-scale real-time nitrogen monitoring in complex orchard environments. To achieve rapid and non-destructive estimation of citrus LNC, this study developed a UAV multispectral inversion framework integrating object-based canopy extraction and machine learning models. Field experiments were conducted in a citrus orchard in western Hubei Province, China. Multi-temporal UAV multispectral images were collected from April to October 2025, and ground measurements of citrus LNC were collected simultaneously. First, minimum distance classification (MDC), maximum likelihood classification (MLC), and object-based image analysis (OBIA) were used for land-cover classification of citrus orchard images, and their canopy extraction performance under complex orchard backgrounds was compared. Subsequently, multiple vegetation indices were calculated from the extracted citrus canopy spectra, and sensitive spectral features were selected through correlation analysis. Finally, seven models, including simple linear regression, quadratic regression, partial least squares regression (PLS), back propagation neural network (BP), extreme learning machine (ELM), particle swarm optimization-extreme learning machine (PSO-ELM), and particle swarm optimization-back propagation neural network (PSO-BP), were constructed to systematically evaluate the inversion performance of citrus LNC across the entire growth period. The results showed that: (1) OBIA achieved higher classification accuracy and temporal stability in citrus orchard land-cover classification, with overall accuracy ranging from 68.86% to 85.65% and Kappa coefficients ranging from 0.56 to 0.72, outperforming MDC and MLC. This indicates that OBIA can effectively reduce the interference of bare soil, grass, shadows, and other non-target objects on canopy spectral extraction. (2) The correlations between vegetation indices and LNC varied markedly among different growth stages, suggesting that the spectral response of citrus LNC has strong phenological dependence and that a single vegetation index is insufficient to stably characterize LNC variation across the whole growth period. (3) At the whole-growth-period scale, multi-index fusion models outperformed single-index models, among which EVI, TVI, and MTVI showed relatively strong cross-stage sensitivity. (4) Optimized machine learning models generally outperformed traditional regression models and unoptimized machine learning models. Among them, PSO-BP achieved the best performance, with a validation R² of 0.68 and an RMSE of 1.54 g kg⁻¹, representing an increase in R² of 23.64% compared with the PLS model and 25.93% over the baseline BP model in terms of R². Overall, this study demonstrates that OBIA-based canopy spectral quality improvement combined with PSO-optimized machine learning can effectively improve the stability and reliability of UAV multispectral estimation of citrus LNC under complex orchard backgrounds. The proposed framework provides technical support for citrus nitrogen diagnosis, precision fertilization, and intelligent orchard management. Full article

(This article belongs to the Topic AI in Optical Spectroscopy Analysis)

►▼ Show Figures

Figure 1

23 pages, 5200 KB  

Open AccessArticle

A Fast Demodulation Algorithm for Fibre Bragg Grating Based on the TimeMixer-LightGBM Hybrid Learning Framework

by Hang Gao, Yizhe Su, Kai Qian, Da Qiu, Song Liu and Tingting Zhang

Sensors 2026, 26(13), 4235; https://doi.org/10.3390/s26134235 - 3 Jul 2026

Viewed by 391

Abstract

Fibre Bragg Grating (FBG) demodulation technology is central to structural health monitoring. However, spectral distortion and noise caused by complex environments, along with the challenge of balancing accuracy and real-time performance in existing deep learning algorithms, severely limit its application in large-scale dynamic [...] Read more.

Fibre Bragg Grating (FBG) demodulation technology is central to structural health monitoring. However, spectral distortion and noise caused by complex environments, along with the challenge of balancing accuracy and real-time performance in existing deep learning algorithms, severely limit its application in large-scale dynamic sensing networks. To address this challenge, this paper proposes a hybrid learning framework named TimeMixer-LightGBM. The framework first employs a pure MLP-based TimeMixer model to efficiently extract multi-scale spectral sequence features from FBG reflection spectra, and then uses a LightGBM model (gradient boosted decision trees) to perform fast regression on these features for centre wavelength shift prediction. Experiments on synthetically distorted FBG spectra (including asymmetric shape variations and additive white noise) show that the method achieves picometre-level accuracy (RMSE = 2.128 pm) with an average processing time of only 0.08 ms per spectrum, representing a speedup of about 4.5 times over the latest deep learning models for FBG demodulation. It also exhibits excellent noise robustness, maintaining an average absolute error of 1.5 pm. Ablation experiments confirm the necessity and synergy of the hybrid architecture. The framework was further applied to demodulate double-peaked overlapping spectra, outperforming existing methods under mild, moderate and severe overlap conditions while keeping inference times in the sub-millisecond range. This study provides a novel and effective technical solution for real-time high-precision FBG demodulation, validates the effectiveness of pure MLP architectures in spectral analysis, and lays a theoretical foundation for deploying such demodulation on embedded edge devices, thereby achieving a favourable balance of accuracy, speed and scalability. Full article

(This article belongs to the Topic AI in Optical Spectroscopy Analysis)

►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less

Select all

Export citation of selected articles as:

Plain Text BibTeX BibTeX (without abstracts) Endnote Endnote (without abstracts) Tab-delimited RIS

 

Error

Oops... you haven't selected anything for export.

Ok

clear

Displaying articles 1-5