Illicit drug production in clandestine laboratories involves the use of large quantities of different chemicals that can be obtained for legitimate purposes. The identification of these chemicals, including reagents, catalyzers and solvents, is crucial for forensic investigations. From a legal point of view, a drug precursor is a material that is specific and critical to the production of a finished chemical and that constitutes a significant portion of the final molecular structure of the drug. In this study, a gas chromatography quartz-enhanced photoacoustic spectroscopy (GC-QEPAS) sensor—in conjunction with a deep learning model—was evaluated for its effectiveness in the detection and identification of interesting compounds for the production of amphetamine, methamphetamine, 3,4-methylenedioxymethamphetamine (MDMA), phenylcyclohexyl piperidine (PCP), and cocaine. The GC-QEPAS sensor includes a gas sampler, a fast GC for separation, and a QEPAS detector, which excites molecules exiting the GC column using a quantum cascade laser to provide the infra-red (IR) spectrum. The on-site capability of the GC-QEPAS system offers significant advantages over the current instruments employed in this field, including rapid analysis, which is crucial in field operations. This allows law enforcement to quickly identify specimens of interest on site. The system’s performance was validated by taking into account the limit of detection, repeatability, and within-laboratory reproducibility. The results showed excellent repeatability and reproducibility for both the GC and QEPAS modules. The deep learning model, a multilayer perceptron neural network, was trained using IR spectra and retention times, achieving very high classification accuracy in the testing conditions. This study demonstrated the efficacy of the GC-QEPAS sensor combined with a deep learning model for the reliable identification of drug precursors, providing a robust tool for law enforcement during criminal investigations in clandestine laboratories.

Achieving fast analytical results on-site with the highest possible accuracy in forensic analyses is crucial for investigations. While portable sensors are essential for crime scene analysis, they often face limitations in sensitivity and specificity, especially due to environmental factors. Data fusion (DF) techniques can enhance accuracy and reliability by combining information from multiple sensors. This study develops different DF approaches using data from two sensors: ion mobility spectrometry (IMS) and gas chromatography-quartz-enhanced photoacoustic spectroscopy (GC-QEPAS), aiming to improve the safety of crime scene operators and the accuracy of on-site forensic analysis. Two DF approaches were developed for acetone and DMMP: low-level (LLDF) and mid-level (MLDF), meanwhile a high-level (HLDF) approach was applied to TATP. LLDF concatenated preprocessed data matrices, while MLDF employed principal component analysis for feature extraction. LLDF and MLDF used one-class support vector machines (OC-SVM) for classification, while HLDF combined OC-SVM for IMS and SIMCA for GC-QEPAS. Sensor location within crime scenes was established using traditional measuring tape and laser distance meters, with a 1 m cutoff distance between sensors deemed appropriate for indoor crime scenes. LLDF achieved high accuracy but was sensitive to concentration variations, while MLDF enhanced the classification robustness. HLDF allowed for independent sensor use in real scenarios. All of the methods reached 100% accuracy for DMMP and acetone, and the MLDF approach was the fastest among the DF methods, demonstrating its potential for rapid applications. DF approaches can significantly enhance the safety and accuracy of forensic investigations, with future research planned to extend data sets and include more sensors.

Artificial Intelligence (AI) is increasingly applied for food quality control, authenticity assessment, and chemical profiling. However, the reliability and industrial applicability of Machine Learning (ML) and Deep Learning (DL) models critically depend on how the datasets are constructed, validated, and interpreted. Among the different analytical techniques used in food analysis, chromatographic fingerprints and chromatographic hyphenated techniques typically generate high-dimensional datasets characterised by structured peak distributions, retention time variability, and strong multicollinearity among adjacent variables, making model development and validation more challenging. These features, combined with limited sample sizes typical of food studies, pose significant challenges for model generalisation and transferability across analytical batches and industrial settings. This work provides a critical evaluation of methodological practices in ML and DL, supported by a systematic review of 165 peer-reviewed studies published between 2015 and 2025. Particular attention is given to common issues such as class imbalance, the misuse of analytical replicates, and the widespread reliance on internal cross-validation without independent test sets. To experimentally assess these factors, a chromatographic dataset on coffee origins was used as a case study to examine how unbalanced class distributions, varying training set sizes, and replicate handling affect model performance. Our results show that high-fold cross-validation and Leave-One-Out Cross-Validation consistently overestimate model accuracy compared to external validation, while treating analytical replicates as independent samples artificially inflates performance without improving true predictive power. Learning curves reveal that most ML models plateau with moderate training sizes, whereas DL models require substantially larger datasets to avoid overfitting. The findings highlight that rigorous validation strategies and careful management of chromatographic data are essential for developing robust, reproducible, and industrially transferable ML models in food chromatography.

Flavour is a key quality attribute of cocoa, essential for industry standards and consumer preferences. Automated methods for assessing flavour quality support industrial laboratories in achieving high sample throughput. Targeted and untargeted HS-SPME-GC-MS chromatographic fingerprints of cocoa volatiles from fermented beans and liquors, combined with machine learning (ML), are used for terroir qualification, enabling effective origin classification with both approaches. The targeted method, which aims to identify chemical patterns associated with sensory attributes is used for flavour comparison of origin with a reference. The similarity analysis successfully identified the most suitable origin to create new blends with a similar flavour to the industry standard. The resulting ML, model based on odorants distribution, enabled the prediction of similarity of blends to the industrial reference with an accuracy of 88 %, a sensitivity of 90 % and a specificity of 84 %.

Background: The detection of explosives in crime scene investigations is critical for forensic science. This study explores the application of laser desorption (LD) ion mobility spectrometry (IMS) as a novel method for this purpose utilising a new IMS prototype developed by MaSaTECH. Methods: The LD sampling technique employs a laser diode module to vaporise explosive traces on surfaces, allowing immediate analysis by IMS without sample preparation. Chemometric approaches, including multivariate data analysis, were utilised for data processing and interpretation, including pre-processing of raw IMS plasmagrams and various pattern recognition techniques, such as linear discriminant analysis (LDA) and support vector machines (SVMs). Results: The IMS prototype was validated through experiments with pure explosives (TNT, RDX, PETN) and explosive products (SEMTEX 1A, C4) on different materials. The study found that the pre-processing method significantly impacts classification accuracy, with the PCA-LDA model demonstrating the best performance for real-world applications. Conclusions: The LD-IMS prototype, coupled with effective chemometric techniques, presents a promising methodology for the detection of explosives in forensic investigations, enhancing the reliability of field applications.

Coffee is a highly complex and variable matrix, with volatile profiles shaped by multiple factors including botanical origin, climatic and soil conditions, post-harvest treatments, and roasting parameters. This variability generates complex chemical patterns, encompassing hundreds of volatile compounds from diverse chemical classes including pyrazines, furans, aldehydes, ketones, and terpenes. The resulting chemical dimensionality poses significant analytical challenges, making the accurate identification of characteristic compounds and reliable discrimination of coffee origins particularly difficult. In this study, we applied comprehensive two-dimensional gas chromatography (GC×GC) coupled with computer vision (CV) to address these challenges. The workflow begins with untargeted fingerprinting, capturing all detectable compounds in a feature template. Multiple sample chromatograms are then combined into composite class images, representing the typical chemical features of each origin while minimising individual variability, which enables rapid pairwise comparison of different origins. CV-based pairwise comparisons highlight differential peaks, which are integrated into a targeted template for subsequent peak extraction. Multivariate analyses then identify the key discriminant compounds driving origin differentiation. Post-processing strategies, such as ion-specific intensity mapping, further enhance interpretability, enabling visualisation of compositional differences across key chemical families. Overall, this GC×GC-CV workflow provides a robust, rapid, and visually intuitive platform for comprehensive chemical characterisation and origin classification of coffee, integrating untargeted and targeted analyses in a single framework.