Recognition of gasoline in fire debris using machine learning: Part I, application of random forest, gradient boosting, support vector machine, and naïve bayes

Assessment of artificial intelligence to detect gasoline in fire debris using HS-SPME-GC/MS and transfer learning

Due to the complex nature of the chemical compositions of ignitable liquids (IL) and the interferences from fire debris matrices, interpreting chromatographic data poses challenges to analysts. In this work, artificial intelligence (AI) was developed by transfer learning in a convolutional neural network (CNN), GoogLeNet. The image classification AI was fine-tuned to create intelligent classification systems to discriminate samples containing gasoline residues from burned substrates. All ground truth samples were analyzed by headspace solid-phase microextraction (HS-SPME) coupled with a gas chromatograph and mass spectrometer (GC/MS). The HS-SPME-GC/MS data were transformed into three types of image presentations, that is, heatmaps, extracted ion heatmaps, and total ion chromatograms. The abundance and mass-to-charge ratios of each scan were converted into image patterns that are characteristic of the chemical profiles of gasoline. The transfer learning data were labeled as "gasoline present" and "gasoline absent" classes. The assessment results demonstrated that all AI models achieved 100 ± 0% accuracy in identifying neat gasoline. When the models were assessed using the spiked samples, the AI model developed using the extracted ion heatmap obtained the highest accuracy rate (95.9 ± 0.4%), which was greater than those obtained by other machine learning models, ranging from 17.3 ± 0.7% to 78.7 ± 0.7%. The proposed work demonstrated that the heatmaps created from GC/MS data can represent chemical features from the samples. Additionally, the pretrained CNN models are readily available in the transfer learning workflow to develop AI for GC/MS data interpretation in fire debris analysis.

Hair fixative traces on footwear - Establishing a link between footwear and the victim's hair after kicks to the head

Kicking a person laying on the floor in the head is a crime whose forensic investigation could profit from additional microtraces capable of linking a suspected footwear, and by extension its owner, to the victim and their injuries. The transfer of hair fixatives (hair gel, hair wax, hair spray, hair foam, etc.) represents such a trace and was consequently practically evaluated throughout this study. This study consists of two parts: The first part, the differentiation study, encompasses the visual, and instrumental analysis of a variety of different hair fixatives to determine their analysability and differentiation potential. The visual examination was conducted using alternate light sources and filter lenses. Subsequently, the instrumental analysis was carried out, whereby the focus lay on Fourier Transform Infra-red (FT-IR) spectroscopy and Raman spectroscopy. The second part is comprised of different experiments including a test-transfer and pendulum experiments to assess the process and the potential variables of the transfer of hair fixative traces between hair and fabric shoes during a kick. This helped to determine the effect of the kick strength and the behaviour of differing hair products. Retrieval methods to secure hair fixative traces of footwear and from the hair of a victim were developed. These were subsequently tested out on an acute case example..

Comparison of decision tree and naïve Bayes algorithms in detecting trace residue of gasoline based on gas chromatography-mass spectrometry data

Fire debris analysis aims to detect and identify any ignitable liquid residues in burnt residues collected at a fire scene. Typically, the burnt residues are analysed using gas chromatography-mass spectrometry (GC-MS) and are manually interpreted. The interpretation process can be laborious due to the complexity and high dimensionality of the GC-MS data. Therefore, this study aims to compare the potential of classification and regression tree (CART) and naïve Bayes (NB) algorithms in analysing the pixel-level GC-MS data of fire debris. The data comprise 14 positive (i.e. fire debris with traces of gasoline) and 24 negative (i.e. fire debris without traces of gasoline) samples. The differences between the positive and negative samples were first inspected based on the mean chromatograms and scores plots of the principal component analysis technique. Then, CART and NB algorithms were independently applied to the GC-MS data. Stratified random resampling was applied to prepare three sets of 200 pairs of training and testing samples (i.e. split ratio of 7:3, 8:2, and 9:1) for estimating the prediction accuracies. Although both the positive and negative samples were hardly differentiated based on the mean chromatograms and scores plots of principal component analysis, the respective NB and CART predictive models produced satisfactory performances with the normalized GC-MS data, i.e. majority achieved prediction accuracy >70%. NB consistently outperformed CART based on the prediction accuracies of testing samples and the corresponding risk of overfitting except when evaluated using only 10% of samples. The accuracy of CART was found to be inversely proportional to the number of testing samples; meanwhile, NB demonstrated rather consistent performances across the three split ratios. In conclusion, NB seems to be much better than CART based on the robustness against the number of testing samples and the consistent lower risk of overfitting.

ATR-FTIR combined with machine learning for the fast non-targeted screening of new psychoactive substances

Due to the diversity and fast evolution of new psychoactive substances (NPS), both public health and safety are threatened around the world. Attenuated total reflection-Fourier transform infrared spectroscopy (ATR-FTIR), which serves as a simple and rapid technique for targeted NPS screening, is challenging with the rapid structural modifications of NPS. To achieve the fast non-targeted screening of NPS, six machine learning (ML) models were constructed to classify eight categories of NPS, including synthetic cannabinoids, synthetic cathinones, phenethylamines, fentanyl analogues, tryptamines, phencyclidine types, benzodiazepines, and "other substances" based on the 1099 IR spectra data items of 362 types of NPS collected by one desktop ATR-FTIR and two portable FTIR spectrometers. All these six ML classification models, including k-nearest neighbor (KNN), support vector machine (SVM), random forest (RF), extra trees (ET), voting, and artificial neural networks (ANNs) were trained through cross validation, and f1-scores of 0.87-1.00 were achieved. In addition, hierarchical cluster analysis (HCA) was performed on 100 synthetic cannabinoids with the most complex structural variation to investigate the structure-spectral property relationship, which leads to a summary of eight synthetic cannabinoid sub-categories with different "linked groups". ML models were also constructed to classify eight synthetic cannabinoid sub-categories. For the first time, this study developed six ML models, which were suitable for both desktop and portable spectrometers, to classify eight categories of NPS and eight synthetic cannabinoids sub-categories. These models can be applied for the fast, accurate, cost-effective, and on-site non-targeted screening of newly emerging NPS with no reference data available.

Fire accelerant classification from GC–MS data of suspected arson cases using machine–learning models

Using a practical GC-MS dataset containing approximately 4000 suspected arson cases, three machine-learning based classification models were developed and their performances were evaluated. All models trained for classifying the data from fire residue into six categories; no fire accelerants detected or else one of fire accelerants was used within gasoline, kerosene, diesel, solvents, or candle. The classification accuracies of the random forest, supporting vector machine, and convolutional neural network model were 0.88, 0.88, and 0.92, respectively. By calculating feature importance of the random forest model, several potential chemical fingerprints of fire accelerants were discovered.

Vulnerable Road Users and Connected Autonomous Vehicles Interaction: A Survey

There is a group of users within the vehicular traffic ecosystem known as Vulnerable Road Users (VRUs). VRUs include pedestrians, cyclists, motorcyclists, among others. On the other hand, connected autonomous vehicles (CAVs) are a set of technologies that combines, on the one hand, communication technologies to stay always ubiquitous connected, and on the other hand, automated technologies to assist or replace the human driver during the driving process. Autonomous vehicles are being visualized as a viable alternative to solve road accidents providing a general safe environment for all the users on the road specifically to the most vulnerable. One of the problems facing autonomous vehicles is to generate mechanisms that facilitate their integration not only within the mobility environment, but also into the road society in a safe and efficient way. In this paper, we analyze and discuss how this integration can take place, reviewing the work that has been developed in recent years in each of the stages of the vehicle-human interaction, analyzing the challenges of vulnerable users and proposing solutions that contribute to solving these challenges.

Recognition of gasoline in fire debris using machine learning: Part II, application of a neural network

The recognition of ignitable liquid (IL) residues in fire debris is a resource intensive but key part of an arson investigation. Due to the highly diverse and heavily loaded chemical matrix of fire debris samples, combined with the broad chemical composition of IL, the interpretation of the laboratory analysis results is a very challenging task for the forensic examiner. Fire debris samples are commonly analyzed using gas chromatography coupled to mass spectrometry (GC-MS). This method delivers both the total ion chromatogram (TIC) with the individually separated compounds and the underlying mass spectrum of each of the separated compounds. In this study, a completely new approach for the recognition of gasoline in fire debris samples is presented. First, the GC-MS data, including retention time, signal intensity, and mass spectrum is converted into a bitmap image. Five different data-to-image conversion approaches are tested, and their advantages and limitations are discussed. Subsequently, a convolutional neural network (CNN) is utilized to allocate the generated images to the classes "with gasoline" or "without gasoline". The applied approaches to generate a digital image and the pattern recognition of the CNN perform very well in the classification of unknown test samples. Depending on the data-to-image generation approach used, the rate of correct sample classification in the test dataset is between 95% and 98%. The machine learning approach in this study, as well as the complementary method presented in an accompanying article, are not only useful for the recognition of gasoline in fire debris but are equally applicable to any additional areas in which the interpretation of complex chromatographic and mass spectrometric is required.