Development of a Detection Algorithm for Use with Reflectance-Based, Real-Time Chemical Sensing

Here, we describe our efforts focused on development of an algorithm for identification of detection events in a real-time sensing application relying on reporting of color values using commercially available color sensing chips. The effort focuses on the identification of event occurrence, rather t...

Full description

Saved in:
Bibliographic Details
Published in:Sensors (Basel, Switzerland) Vol. 16; no. 11; p. 1927
Main Authors: Malanoski, Anthony P, Johnson, Brandy J, Erickson, Jeffrey S, Stenger, David A
Format: Journal Article
Language:English
Published: Switzerland MDPI AG 16-11-2016
MDPI
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Here, we describe our efforts focused on development of an algorithm for identification of detection events in a real-time sensing application relying on reporting of color values using commercially available color sensing chips. The effort focuses on the identification of event occurrence, rather than target identification, and utilizes approaches suitable to onboard device incorporation to facilitate portable and autonomous use. The described algorithm first excludes electronic noise generated by the sensor system and determines response thresholds. This automatic adjustment provides the potential for use with device variations as well as accommodating differing indicator behaviors. Multiple signal channels (RGB) as well as multiple indicator array elements are combined for reporting of an event with a minimum of false responses. While the method reported was developed for use with paper-supported porphyrin and metalloporphyrin indicators, it should be equally applicable to other colorimetric indicators. Depending on device configurations, receiver operating characteristic (ROC) sensitivities of 1 could be obtained with specificities of 0.87 (threshold 160 ppb, ethanol).
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1424-8220
1424-8220
DOI:10.3390/s16111927