ENHANCE CATEGORISATION OF MULTILEVEL HIGH-SENSITIVITY CARDIOVASCULAR BIOMARKERS FROM LATERAL FLOW IMMUNOASSAY IMAGES VIA NEURAL NETWORKS AND DYNAMIC TIME WARPING

Min Jing, Brian Mac Namee, Donal Mc Laughlin, David Steele, Sara Mc Namee, Patrick Cullen, Dewar Finlay, James McLaughlin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

30 Downloads (Pure)

Abstract

Lateral Flow Immunoassays (LFA) are low cost, rapid and highly efficacious Point-of-Care devices. Traditional LFA testing faces challenges to detect high-sensitivity biomarkers due to low sensitivity. Most approaches are based on averaging image intensity from a region-of-interest (ROI). This paper presents a novel system that considers each row of an LFA image as a time series signal and, consequently, does not require the detection of ROI. Long Short-Term Memory (LSTM) networks are used to classify LFA data obtained from multiscale high-sensitivity cardiovascular biomarkers. Dynamic Time Warping (DTW) was incorporated with LSTM to align the LFA data from different concentration levels to a common reference before feeding the distance maps into an LSTM network. The LSTM network outperforms other classifiers with or without DTW. Furthermore, performance of all classifiers is improved after incorporating DTW. The positive outcomes suggest the potential of the proposed system for early risk assessment of cardiovascular diseases.
Original languageEnglish
Title of host publicationThe Proceedings of IEEE ICIP 2020
PublisherIEEE
Publication statusAccepted/In press - 16 May 2020
EventThe 27th IEEE International Conference on Image Processing (ICIP 2020) -
Duration: 25 Oct 202028 Oct 2020
https://2020.ieeeicip.org/

Conference

ConferenceThe 27th IEEE International Conference on Image Processing (ICIP 2020)
Abbreviated titleICIP2020
Period25/10/2028/10/20
Internet address

Fingerprint Dive into the research topics of 'ENHANCE CATEGORISATION OF MULTILEVEL HIGH-SENSITIVITY CARDIOVASCULAR BIOMARKERS FROM LATERAL FLOW IMMUNOASSAY IMAGES VIA NEURAL NETWORKS AND DYNAMIC TIME WARPING'. Together they form a unique fingerprint.

Cite this