Methodology for the Implementation of Internal Standard to Laser-Induced Breakdown Spectroscopy Analysis of Soft Tissues.

The improving performance of the laser-induced breakdown spectroscopy (LIBS) triggered its utilization in the challenging topic of soft tissue analysis. Alterations of elemental content within soft tissues are commonly assessed and provide further insights in biological research. However, the laser ablation of soft tissues is a complex issue and demands a priori optimization, which is not straightforward in respect to a typical LIBS experiment. Here, we focus on implementing an internal standard into the LIBS elemental analysis of soft tissue samples. We achieve this by extending routine methodology for optimization of soft tissues analysis with a standard spiking method. This step enables a robust optimization procedure of LIBS experimental settings. Considering the implementation of LIBS analysis to the histological routine, we avoid further alterations of the tissue structure. Therefore, we propose a unique methodology of sample preparation, analysis, and subsequent data treatment, which enables the comparison of signal response from heterogenous matrix for different LIBS parameters. Additionally, a brief step-by-step process of optimization to achieve the highest signal-to-noise ratio (SNR) is described. The quality of laser-tissue interaction is investigated on the basis of the zinc signal response, while selected experimental parameters (e.g., defocus, gate delay, laser energy, and ambient atmosphere) are systematically modified.

Analysis of Laser-Induced Breakdown Spectroscopy Data Acquired from Boundary of Two Matrices.

Laser-induced breakdown spectroscopy (LIBS) data obtained from the elemental imaging of heterogenous samples were processed with various chemometric algorithms. The intention was to cluster obtained characteristic spectra and to provide additional information about the sample surface composition and distribution of individual matrices. However, there is a gray zone on the boundary of two matrices and the consequent clustering of the spectra obtained on this boundary is ambiguous. This paper focuses on the transition between two well-defined matrices in a simplified case for a better transparency in data visualization. Steel and aluminum samples that are represented by characteristic spectra with significantly distinct structures (e.g., different number of spectral lines). Using a carefully designed experiment, several Fe:Al ratios were ablated and analyzed by principal component analysis (PCA), self-organizing maps (SOM), and standard data metrics. This paper shows the strategy for the discrimination of unrecognized spectra and possibilities in their clustering.

Interpreting support vector machines applied in laser-induced breakdown spectroscopy.

Laser-induced breakdown spectroscopy is often combined with a multivariate black box model-such as support vector machines (SVMs)-to obtain desirable quantitative or qualitative results. This approach carries obvious risks when practiced in high-stakes applications. Moreover, the lack of understanding of a black-box model limits the user's ability to fine-tune the model. Thus, here we present four approaches to interpret SVMs through investigating which features the models consider important in the classification task of 19 algal and cyanobacterial species. The four feature importance metrics are compared with popular approaches to feature selection for optimal SVM performance. We report that the distinct feature importance metrics yield complementary and often comparable information. In addition, we identify our SVM model's bias towards features with a large variance, even though these features exhibit a significant overlap between classes. We also show that the linear and radial basis kernel SVMs weight the same features to the same degree.

Benchmark classification dataset for laser-induced breakdown spectroscopy.

In this work, we present an extensive dataset of laser-induced breakdown spectroscopy (LIBS) spectra for the pre-training and evaluation of LIBS classification models. LIBS is a well-established spectroscopic method for in-situ and industrial applications, where LIBS is primarily applied for clustering and classification tasks. As such, our dataset is aimed at helping with the development and testing of classification and clustering methodologies. Moreover, the dataset could be used to pre-train classification models for applications where the amount of available data is limited. The dataset consists of LIBS spectra of 138 soil samples belonging to 12 distinct classes. The spectra were acquired with a state-of-the-art LIBS system. Lastly, the composition of each sample is also provided, including estimated uncertainties.