In recent years, chemometric applications have expanded across multiple scientific domains, showing remarkable growth and diversification. This development is largely due to the integration of chemometric methods with spectroscopic techniques, which offer rapid, non-destructive, and cost-effective analytical solutions. In forensic science, and particularly in micro-trace analysis, spectroscopy coupled with chemometrics has proven invaluable for interpreting complex data and ensuring result repeatability even from minimal evidence. However, the absence of standardized reference databases—unlike those available in genetic investigations—prevents the assignment of quantitative probabilistic values to analytical outcomes. Consequently, forensic experts are often limited to expressing qualitative opinions regarding evidential support, creating challenges for judicial interpretation and communication. This research aims to bridge the gap between analytical results and their forensic–judicial interpretation by evaluating which chemometric classification methods—LDA, PLS-DA, SIMCA, or KNN—are most suitable for expressing evidential value in court. Building upon the conceptual framework of the Sydney Declaration (2022), which redefines the principles of forensic science, the project integrates real forensic case data with experimental chemometric analyses conducted at the Microanalysis Laboratory of the Scientific Investigation Department (RIS) in Rome. The study also explores the potential for probabilistic, data-driven classification models to enhance transparency, reproducibility, and clarity in courtroom communication. Additional activities include the application of chemometric techniques to multispectral imaging within the H2020 “RISEN” project (Real-tIme on-site forenSic tracE qualificatioN) and collaborative research on the forensic characterization of narcotic substances and adulterants through near-infrared imaging, presented at NIR Italia 2025 (Rome, June 2024). Ultimately, this work contributes to strengthening the interpretative and communicative reliability of scientific evidence in judicial contexts, aligning forensic analytical practices with the ethical and methodological principles outlined in the Sydney Declaration.

Application of chemometric techniques in the field of forensic investigations

CASAMASSIMA, ROSARIO
2026

Abstract

In recent years, chemometric applications have expanded across multiple scientific domains, showing remarkable growth and diversification. This development is largely due to the integration of chemometric methods with spectroscopic techniques, which offer rapid, non-destructive, and cost-effective analytical solutions. In forensic science, and particularly in micro-trace analysis, spectroscopy coupled with chemometrics has proven invaluable for interpreting complex data and ensuring result repeatability even from minimal evidence. However, the absence of standardized reference databases—unlike those available in genetic investigations—prevents the assignment of quantitative probabilistic values to analytical outcomes. Consequently, forensic experts are often limited to expressing qualitative opinions regarding evidential support, creating challenges for judicial interpretation and communication. This research aims to bridge the gap between analytical results and their forensic–judicial interpretation by evaluating which chemometric classification methods—LDA, PLS-DA, SIMCA, or KNN—are most suitable for expressing evidential value in court. Building upon the conceptual framework of the Sydney Declaration (2022), which redefines the principles of forensic science, the project integrates real forensic case data with experimental chemometric analyses conducted at the Microanalysis Laboratory of the Scientific Investigation Department (RIS) in Rome. The study also explores the potential for probabilistic, data-driven classification models to enhance transparency, reproducibility, and clarity in courtroom communication. Additional activities include the application of chemometric techniques to multispectral imaging within the H2020 “RISEN” project (Real-tIme on-site forenSic tracE qualificatioN) and collaborative research on the forensic characterization of narcotic substances and adulterants through near-infrared imaging, presented at NIR Italia 2025 (Rome, June 2024). Ultimately, this work contributes to strengthening the interpretative and communicative reliability of scientific evidence in judicial contexts, aligning forensic analytical practices with the ethical and methodological principles outlined in the Sydney Declaration.
13-gen-2026
Inglese
MARINI, Federico
Università degli Studi di Roma "La Sapienza"
144
File in questo prodotto:
File Dimensione Formato  
Tesi_dottorato_Casamassima.pdf

accesso aperto

Licenza: Creative Commons
Dimensione 3.63 MB
Formato Adobe PDF
3.63 MB Adobe PDF Visualizza/Apri

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14242/366391
Il codice NBN di questa tesi è URN:NBN:IT:UNIROMA1-366391