Digital Fingerprinting of Complex Liquids Using a Reconfigurable Multi‐Sensor System with Foundation Models

Abstract Combining chemical sensor arrays with machine learning enables designing intelligent systems to perform complex sensing tasks and unveil properties that are not directly accessible through conventional analytical chemistry. However, personalized and portable sensor systems are typically uns...

Full description

Saved in:
Bibliographic Details
Main Authors: Gianmarco Gabrieli, Matteo Manica, Joris Cadow‐Gossweiler, Patrick W. Ruch
Format: Article
Language:English
Published: Wiley 2024-11-01
Series:Advanced Science
Subjects:
Online Access:https://doi.org/10.1002/advs.202407513
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850146040006574080
author Gianmarco Gabrieli
Matteo Manica
Joris Cadow‐Gossweiler
Patrick W. Ruch
author_facet Gianmarco Gabrieli
Matteo Manica
Joris Cadow‐Gossweiler
Patrick W. Ruch
author_sort Gianmarco Gabrieli
collection DOAJ
description Abstract Combining chemical sensor arrays with machine learning enables designing intelligent systems to perform complex sensing tasks and unveil properties that are not directly accessible through conventional analytical chemistry. However, personalized and portable sensor systems are typically unsuitable for the generation of extensive data sets, thereby limiting the ability to train large models in the chemical sensing realm. Foundation models have demonstrated unprecedented zero‐shot learning capabilities on various data structures and modalities, in particular for language and vision. Transfer learning from such models is explored by providing a framework to create effective data representations for chemical sensors and ultimately describe a novel, generalizable approach for AI‐assisted chemical sensing. The translation of signals produced by remarkably simple and portable multi‐sensor systems into visual fingerprints of liquid samples under test is demonstrated, and it is illustrated that how a pipeline incorporating pretrained vision models yields >95% average classification accuracy in four unrelated chemical sensing tasks with limited domain‐specific training measurements. This approach matches or outperforms expert‐curated sensor signal features, thereby providing a generalization of data processing for ultimate ease‐of‐use and broad applicability to enable interpretation of multi‐signal outputs for generic sensing applications.
format Article
id doaj-art-4b53d36babc04c8d8ea3f673d0a78006
institution OA Journals
issn 2198-3844
language English
publishDate 2024-11-01
publisher Wiley
record_format Article
series Advanced Science
spelling doaj-art-4b53d36babc04c8d8ea3f673d0a780062025-08-20T02:27:57ZengWileyAdvanced Science2198-38442024-11-011144n/an/a10.1002/advs.202407513Digital Fingerprinting of Complex Liquids Using a Reconfigurable Multi‐Sensor System with Foundation ModelsGianmarco Gabrieli0Matteo Manica1Joris Cadow‐Gossweiler2Patrick W. Ruch3IBM Research Europe Säumerstrasse 4 Rüschlikon 8803 SwitzerlandIBM Research Europe Säumerstrasse 4 Rüschlikon 8803 SwitzerlandIBM Research Europe Säumerstrasse 4 Rüschlikon 8803 SwitzerlandIBM Research Europe Säumerstrasse 4 Rüschlikon 8803 SwitzerlandAbstract Combining chemical sensor arrays with machine learning enables designing intelligent systems to perform complex sensing tasks and unveil properties that are not directly accessible through conventional analytical chemistry. However, personalized and portable sensor systems are typically unsuitable for the generation of extensive data sets, thereby limiting the ability to train large models in the chemical sensing realm. Foundation models have demonstrated unprecedented zero‐shot learning capabilities on various data structures and modalities, in particular for language and vision. Transfer learning from such models is explored by providing a framework to create effective data representations for chemical sensors and ultimately describe a novel, generalizable approach for AI‐assisted chemical sensing. The translation of signals produced by remarkably simple and portable multi‐sensor systems into visual fingerprints of liquid samples under test is demonstrated, and it is illustrated that how a pipeline incorporating pretrained vision models yields >95% average classification accuracy in four unrelated chemical sensing tasks with limited domain‐specific training measurements. This approach matches or outperforms expert‐curated sensor signal features, thereby providing a generalization of data processing for ultimate ease‐of‐use and broad applicability to enable interpretation of multi‐signal outputs for generic sensing applications.https://doi.org/10.1002/advs.202407513chemical sensingmulti‐sensor systemsvision foundation models
spellingShingle Gianmarco Gabrieli
Matteo Manica
Joris Cadow‐Gossweiler
Patrick W. Ruch
Digital Fingerprinting of Complex Liquids Using a Reconfigurable Multi‐Sensor System with Foundation Models
Advanced Science
chemical sensing
multi‐sensor systems
vision foundation models
title Digital Fingerprinting of Complex Liquids Using a Reconfigurable Multi‐Sensor System with Foundation Models
title_full Digital Fingerprinting of Complex Liquids Using a Reconfigurable Multi‐Sensor System with Foundation Models
title_fullStr Digital Fingerprinting of Complex Liquids Using a Reconfigurable Multi‐Sensor System with Foundation Models
title_full_unstemmed Digital Fingerprinting of Complex Liquids Using a Reconfigurable Multi‐Sensor System with Foundation Models
title_short Digital Fingerprinting of Complex Liquids Using a Reconfigurable Multi‐Sensor System with Foundation Models
title_sort digital fingerprinting of complex liquids using a reconfigurable multi sensor system with foundation models
topic chemical sensing
multi‐sensor systems
vision foundation models
url https://doi.org/10.1002/advs.202407513
work_keys_str_mv AT gianmarcogabrieli digitalfingerprintingofcomplexliquidsusingareconfigurablemultisensorsystemwithfoundationmodels
AT matteomanica digitalfingerprintingofcomplexliquidsusingareconfigurablemultisensorsystemwithfoundationmodels
AT joriscadowgossweiler digitalfingerprintingofcomplexliquidsusingareconfigurablemultisensorsystemwithfoundationmodels
AT patrickwruch digitalfingerprintingofcomplexliquidsusingareconfigurablemultisensorsystemwithfoundationmodels