Determining time since deposition of latent fingerprints on forensic adhesive tape using ultrafast DESI-MS and machine learning

Abstract Fingerprints provide indisputable forensic evidence for establishing identity. Latent fingerprints, often visualized with black magnetic powder and recovered with adhesive tape, can be matched to police databases for identification. However, determining the time since deposition (TSD) is cr...

Full description

Saved in:
Bibliographic Details
Main Authors: Nora Rajs, Yinon Harush-Brosh, Ron Raisch, Ravit Yakobi Arancibia, Amani Zoabi, Guy Nevet Golan, Moshe Shpitzen, Sarena Wiesner, Michal Levin-Elad, Tommy Kaplan, Katherine Margulis
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-025-02639-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849688406031859712
author Nora Rajs
Yinon Harush-Brosh
Ron Raisch
Ravit Yakobi Arancibia
Amani Zoabi
Guy Nevet Golan
Moshe Shpitzen
Sarena Wiesner
Michal Levin-Elad
Tommy Kaplan
Katherine Margulis
author_facet Nora Rajs
Yinon Harush-Brosh
Ron Raisch
Ravit Yakobi Arancibia
Amani Zoabi
Guy Nevet Golan
Moshe Shpitzen
Sarena Wiesner
Michal Levin-Elad
Tommy Kaplan
Katherine Margulis
author_sort Nora Rajs
collection DOAJ
description Abstract Fingerprints provide indisputable forensic evidence for establishing identity. Latent fingerprints, often visualized with black magnetic powder and recovered with adhesive tape, can be matched to police databases for identification. However, determining the time since deposition (TSD) is crucial to temporally tie the fingerprints to the crime. Despite extensive efforts, no reliable method exists for determining TSD. This study presents a workflow for directly dating fingerprints using ultrafast 2-dimensional desorption electrospray ionization mass spectrometry (DESI-MS). The fingerprints are analyzed directly from a forensic tape after development with magnetic powder. This method aims to enable dating of fingerprints collected from virtually any non-porous surface. The study involved 744 fingerprints from 330 volunteers, aged up to 15 days under various conditions. Data analysis using the XGBoost and SMOTE algorithms achieved a correlation of 0.54 (p-value < 1e−5) between TSD prediction and true TSD, achieving 83.3% accuracy in distinguishing between 0-4 days and 10–15 days old prints. Key imaging parameters, such as DESI-MS scan rate, mass range, scan area, spatial resolution, and imaging mode, were optimized to enhance age determination precision and support rapid processing within forensic workflows. This research, conducted in collaboration between police forensic units and an academic institution, integrates seamlessly into practical forensic applications.
format Article
id doaj-art-e5ed4f09389f49f6b39925dea5de24e5
institution DOAJ
issn 2045-2322
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-e5ed4f09389f49f6b39925dea5de24e52025-08-20T03:22:02ZengNature PortfolioScientific Reports2045-23222025-05-0115111210.1038/s41598-025-02639-yDetermining time since deposition of latent fingerprints on forensic adhesive tape using ultrafast DESI-MS and machine learningNora Rajs0Yinon Harush-Brosh1Ron Raisch2Ravit Yakobi Arancibia3Amani Zoabi4Guy Nevet Golan5Moshe Shpitzen6Sarena Wiesner7Michal Levin-Elad8Tommy Kaplan9Katherine Margulis10Latent Fingerprint Laboratory, Division of Identification and Forensic Science (DIFS), Israel PoliceR&D department, Division of Identification and Forensic Science (DIFS), Israel PoliceR&D department, Division of Identification and Forensic Science (DIFS), Israel PoliceThe Institute for Drug Research, School of Pharmacy, Faculty of Medicine, The Hebrew University of JerusalemThe Institute for Drug Research, School of Pharmacy, Faculty of Medicine, The Hebrew University of JerusalemR&D department, Division of Identification and Forensic Science (DIFS), Israel PoliceR&D department, Division of Identification and Forensic Science (DIFS), Israel PoliceLatent Fingerprint Laboratory, Division of Identification and Forensic Science (DIFS), Israel PoliceDivision of Identification and Forensic Science (DIFS), Department of National Forensic Investigations, National H.Q. JerusalemSchool of Computer Science and Engineering, The Hebrew University of JerusalemThe Institute for Drug Research, School of Pharmacy, Faculty of Medicine, The Hebrew University of JerusalemAbstract Fingerprints provide indisputable forensic evidence for establishing identity. Latent fingerprints, often visualized with black magnetic powder and recovered with adhesive tape, can be matched to police databases for identification. However, determining the time since deposition (TSD) is crucial to temporally tie the fingerprints to the crime. Despite extensive efforts, no reliable method exists for determining TSD. This study presents a workflow for directly dating fingerprints using ultrafast 2-dimensional desorption electrospray ionization mass spectrometry (DESI-MS). The fingerprints are analyzed directly from a forensic tape after development with magnetic powder. This method aims to enable dating of fingerprints collected from virtually any non-porous surface. The study involved 744 fingerprints from 330 volunteers, aged up to 15 days under various conditions. Data analysis using the XGBoost and SMOTE algorithms achieved a correlation of 0.54 (p-value < 1e−5) between TSD prediction and true TSD, achieving 83.3% accuracy in distinguishing between 0-4 days and 10–15 days old prints. Key imaging parameters, such as DESI-MS scan rate, mass range, scan area, spatial resolution, and imaging mode, were optimized to enhance age determination precision and support rapid processing within forensic workflows. This research, conducted in collaboration between police forensic units and an academic institution, integrates seamlessly into practical forensic applications.https://doi.org/10.1038/s41598-025-02639-y
spellingShingle Nora Rajs
Yinon Harush-Brosh
Ron Raisch
Ravit Yakobi Arancibia
Amani Zoabi
Guy Nevet Golan
Moshe Shpitzen
Sarena Wiesner
Michal Levin-Elad
Tommy Kaplan
Katherine Margulis
Determining time since deposition of latent fingerprints on forensic adhesive tape using ultrafast DESI-MS and machine learning
Scientific Reports
title Determining time since deposition of latent fingerprints on forensic adhesive tape using ultrafast DESI-MS and machine learning
title_full Determining time since deposition of latent fingerprints on forensic adhesive tape using ultrafast DESI-MS and machine learning
title_fullStr Determining time since deposition of latent fingerprints on forensic adhesive tape using ultrafast DESI-MS and machine learning
title_full_unstemmed Determining time since deposition of latent fingerprints on forensic adhesive tape using ultrafast DESI-MS and machine learning
title_short Determining time since deposition of latent fingerprints on forensic adhesive tape using ultrafast DESI-MS and machine learning
title_sort determining time since deposition of latent fingerprints on forensic adhesive tape using ultrafast desi ms and machine learning
url https://doi.org/10.1038/s41598-025-02639-y
work_keys_str_mv AT norarajs determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT yinonharushbrosh determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT ronraisch determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT ravityakobiarancibia determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT amanizoabi determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT guynevetgolan determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT mosheshpitzen determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT sarenawiesner determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT michallevinelad determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT tommykaplan determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning
AT katherinemargulis determiningtimesincedepositionoflatentfingerprintsonforensicadhesivetapeusingultrafastdesimsandmachinelearning