Chemometrics in Forensic Chemistry - Part III: Quality assessment and interpretation of chemometric output.

The number of samples sent to forensic laboratories as well as the complexity of the drug situation has increased tremendously during recent years. At the same time the amount of data gathered from chemical measurements has been mounting. This creates challenges for forensic chemists: how to handle the data, how to reliably answer the questions asked, and how to examine the data to find new properties or how to disclose connections with respect to source attribution of samples within a case or retrospective to past cases, stored in a database. Previously published articles Chemometrics in Forensic Chemistry - Part I and II discussed where in the forensic workflow of routine casework chemometrics is applied, and presented examples of chemometric methods used in cases of illicit drugs. This article explains through examples that the chemometric results must never stand-alone. Before such results are reported, quality assessment steps, which may consist of operational, chemical, and forensic assessments are required. In each case a forensic chemist needs to consider the suitability of chemometric methods, based on their strengths, weaknesses, opportunities and threats (SWOT). This is because while chemometric methods are powerful tools managing complex data, they are to some extent chemically blind.

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.

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

The detection and identification of ignitable liquid (IL) residues in fire debris are two very challenging tasks in a fire investigation. To this day, the recognition of IL in fire debris includes the chemical analysis of the fire debris composition, followed by the examination and interpretation of the analysis result by a trained forensic examiner. Throughout the last decade, chemometrics and artificial intelligence have become increasingly important. In the present study, machine learning algorithms capable of recognizing gasoline residues in fire debris based on GC-MS data have been developed. Four methods, including random forest, gradient boosting, support vector machine, and naïve bayes are applied and used to classify fire debris samples into the two categories "with gasoline" or "without gasoline". A fifth method (logistic regression) did not converge due to well separated classes. A database comprising 360 measurements, including fire debris samples of real cases as well as fire debris samples spiked with known amounts of weathered gasoline (up to 99.6%), was available to train the machine learning algorithms (using 85% of the data) and to subsequently test the performance of the methods when classifying unknown samples (using 15% of the data). In general, the methods perform very well, as three of it succeeded to classify all test samples correctly without any false positive or false negative allocations. One (naïve bayes) was not trained enough to classify other (non-gasoline) IL correctly as "no gasoline". Furthermore, the random forest method reveals which chemical compounds are most relevant for the algorithm to classify the samples. In general, the presented approach is highly promising and could easily be extended or adapted to other types of IL. Similar to the neural network presented in the accompanying paper, such methods have the potential to serve as a fast screening technique for fire debris samples, thus supporting the forensic examiner by providing an additional independent opinion. Nonetheless, the definite identification of IL residues in fire debris always has to be accomplished by a forensic examiner.

Chemometrics in forensic chemistry - Part II: Standardized applications - Three examples involving illicit drugs.

In the recently published article "Chemometrics in forensic chemistry - Part I: Implications to the forensic workflow" the application of chemometric methods in forensic casework was described. The steps to facilitate standardized chemometric procedures and the availability of chemometric tools such as software and a guideline are under development. Three examples of typical illicit drugs casework, wherein chemometric methods were applied, are presented in the current paper. The kind of questions presented in these examples cover identification, classification, comparison and quantification of illicit drugs. The examples include several types of data (low- or high-dimensional), pre-processing and chemometric analyses that are applied to answer the questions presented. The performance measures for the chemometric methods are described based on separate datasets for training and testing (validation) purposes. In this way it is illustrated how a chemometric method is set up and data analysis may be performed. The presented methods are intended to be easily translatable to questions in forensic chemistry that are not drug-related.

Chemometrics in forensic chemistry - Part I: Implications to the forensic workflow.

The forensic literature shows a clear trend towards increasing use of chemometrics (i.e. multivariate analysis and other statistical methods). This can be seen in different disciplines such as drug profiling, arson debris analysis, spectral imaging, glass analysis, age determination, and more. In particular, current chemometric applications cover low-dimensional (e.g. drug impurity profiles) and high-dimensional data (e.g. Infrared and Raman spectra) and are therefore useful in many forensic disciplines. There is a dominant and increasing need in forensic chemistry for reliable and structured processing and interpretation of analytical data. This is especially true when classification (grouping) or profiling (batch comparison) is of interest. Chemometrics can provide additional information in complex crime cases and enhance productivity by improving the processes of data handling and interpretation in various applications. However, the use of chemometrics in everyday work tasks is often considered demanding by forensic scientists and, consequently, they are only reluctantly used. This article and following planned contributions are dedicated to those forensic chemists, interested in applying chemometrics but for any reasons are limited in the proper application of statistical tools - usually made for professionals - or the direct support of statisticians. Without claiming to be comprehensive, the literature reviewed revealed a sufficient overview towards the preferably used data handling and chemometric methods used to answer the forensic question. With this basis, a software tool will be designed (part of the EU project STEFA-G02) and handed out to forensic chemist with all necessary elements of data handling and evaluation. Because practical casework is less and less accompanied from the beginning to the end out of the same hand, more and more interfaces are built in through specialization of individuals. This article presents key influencing elements in the forensic workflow related to the most meaningful chemometric application and evaluation.

Sampling of illicit drugs for quantitative analysis--part III: sampling plans and sample preparations.

The findings in this paper are based on the results of our drug homogeneity studies and particle size investigations. Using that information, a general sampling plan (depicted in the form of a flow-chart) was devised that could be applied to the quantitative instrumental analysis of the most common illicit drugs: namely heroin, cocaine, amphetamine, cannabis resin, MDMA tablets and herbal cannabis in 'bud' form (type I). Other more heterogeneous forms of cannabis (type II) were found to require alternative, more traditional sampling methods. A table was constructed which shows the sampling uncertainty expected when a particular number of random increments are taken and combined to form a single primary sample. It also includes a recommended increment size; which is 1 g for powdered drugs and cannabis resin, 1 tablet for MDMA and 1 bud for herbal cannabis in bud form (type I). By referring to that table, individual laboratories can ensure that the sampling uncertainty for a particular drug seizure can be minimised, such that it lies in the same region as their analytical uncertainty for that drug. The table shows that assuming a laboratory wishes to quantitatively analyse a seizure of powdered drug or cannabis resin with a 'typical' heterogeneity, a primary sample of 15×1 g increments is generally appropriate. The appropriate primary sample for MDMA tablets is 20 tablets, while for herbal cannabis (in bud form) 50 buds were found to be appropriate. Our study also showed that, for a suitably homogenised primary sample of the most common powdered drugs, an analytical sample size of between 20 and 35 mg was appropriate and for herbal cannabis the appropriate amount was 200 mg. The need to ensure that the results from duplicate or multiple incremental sampling were compared, to demonstrate whether or not a particular seized material has a 'typical' heterogeneity and that the sampling procedure applied has resulted in a 'correct sample', was highlighted and the setting up of suitable control charts (R or S charts), for quality control purposes, was strongly recommended and examples given. Furthermore, although this particular study relates to the sampling of illicit drugs, it should be remembered that it is based on general sampling theory and therefore the same approach could be applied to other disciplines where 'correct sampling' of powders and solids is important.

Sampling of illicit drugs for quantitative analysis--part II. Study of particle size and its influence on mass reduction.

The basic goal in sampling for the quantitative analysis of illicit drugs is to maintain the average concentration of the drug in the material from its original seized state (the primary sample) all the way through to the analytical sample, where the effect of particle size is most critical. The size of the largest particles of different authentic illicit drug materials, in their original state and after homogenisation, using manual or mechanical procedures, was measured using a microscope with a camera attachment. The comminution methods employed included pestle and mortar (manual) and various ball and knife mills (mechanical). The drugs investigated were amphetamine, heroin, cocaine and herbal cannabis. It was shown that comminution of illicit drug materials using these techniques reduces the nominal particle size from approximately 600 µm down to between 200 and 300 µm. It was demonstrated that the choice of 1 g increments for the primary samples of powdered drugs and cannabis resin, which were used in the heterogeneity part of our study (Part I) was correct for the routine quantitative analysis of illicit seized drugs. For herbal cannabis we found that the appropriate increment size was larger. Based on the results of this study we can generally state that: An analytical sample weight of between 20 and 35 mg of an illicit powdered drug, with an assumed purity of 5% or higher, would be considered appropriate and would generate an RSDsampling in the same region as the RSDanalysis for a typical quantitative method of analysis for the most common, powdered, illicit drugs. For herbal cannabis, with an assumed purity of 1% THC (tetrahydrocannabinol) or higher, an analytical sample weight of approximately 200 mg would be appropriate. In Part III we will pull together our homogeneity studies and particle size investigations and use them to devise sampling plans and sample preparations suitable for the quantitative instrumental analysis of the most common illicit drugs.

Sampling of illicit drugs for quantitative analysis. Part I: heterogeneity study of illicit drugs in Europe.

Sampling of illicit drugs for qualitative and quantitative analysis would normally be considered as routine and comparable tasks in forensic drugs laboratories and previously similar statistical sampling approaches have been applied. However, we believe that two different sampling approaches, based on two different theoretical and statistical backgrounds are more appropriate. Furthermore the application of the qualitative sampling approach can be impractical for quantitative sampling as it could generate many analytical samples from a single seizure. In some countries the purity of the illicit drug in a seizure may affect the criminal sentence and therefore, reliable results for quantitative analysis are crucial. It was decided to investigate a new approach, which although incorporating some statistics also took account of our background knowledge about the composition of the drugs we were analysing. The ultimate goal was to produce recommendations for a practical sampling plan for quantitative analysis. It was found that the two key factors which had a significant effect on obtaining a representative analytical sample from a bulk seizure were the heterogeneity of the drug powder and the particle sizes of its components. This article concentrates on drug heterogeneity. Particle size effects will be addressed in part II of this study. A sampling plan was devised for a range of drug seizure types and asked ENFSI member laboratories to use it when analysing real drug seizures to provide heterogeneity data for the most common illicit drugs (heroin, cocaine, amphetamine, MDMA and cannabis (herbal and resin)). It was found that for routine quantitative drugs analysis, the sampling problems caused by heterogeneity can be solved by using an incremental sampling protocol. Furthermore, the number of increments that need to be taken for a particular drug is dependent on the relative standard deviation (RSD) required by an individual laboratory and the analytical method that they employ. A 1g increment size was found to be suitable for powdered drugs and cannabis resin. However, 1g increments were not suitable for herbal cannabis, because of particle size issues. Sampling of herbal cannabis will be addressed in Part II of this study. Recommendations for a sampling plan, based on the heterogeneity and particle size of specific drugs seizures in casework will be discussed in Part III of this study.

Ketogenic diet effects on cognition, mood, and psychosocial adjustment in children.

INTRODUCTION: The ketogenic diet (KD) is increasingly used for the treatment of refractory epilepsy. The aim of this study was to evaluate possible adverse effects of the diet on cognition, behavior, psychosocial adjustment, and quality of life in school-aged children and adolescents. METHOD: Fifteen subjects were assessed before diet initiation. After approximately 6 months, on diet treatment 11 patients (73%) were reassessed. We used a combination of individually administered psychological tests for the children and parent report questionnaires. RESULTS: Five of 15 patients had a seizure reduction of more than 50%. Cognition showed a small trend toward improvement in most patients. Psychosocial adjustment, on the other hand, showed small trends toward worsening. For mood, two areas showed a larger change, revealing more mood problems although this was not on a statistically significant level. CONCLUSION: In this small group of children, there is no indication that the KD has a negative impact on cognition or social adaptation at short term. There is a tendency toward an increase in mood problems.