Non-Invasive Blood Group Detection Using Photoplethysmography And Deep Learning

Read full text Summarise with VIRAT
Authors: S Hemant Kumar, T. Prasanna Lakshmi, M Anil Reddy, K Suresh Babu
Cited by: 0
References: 4

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Traditionalmethodsofidentifyingbloodgroupsinvolvesamplecollectionbyveinousbloodofindividuals,whichinvolvesanel ementofinfections,involves qualified human resources, and presents time setbacks which may be fatal in acute clinical settings.In this manuscript, a non-invasivemethod of blood group identification is presented that consists of combining photoplethysmographic (PPG) signal acquisition with a deepconvoluted neural network classifier.Raw PPG signals are obtained through an optical sensor based on fingers with wavelength of 940nm; then,morphologicalparametersincludingpulsetransittime,dicroticnotchamplitude,slopecoefficientsofthewaveformareobtainedandmadeavailabletoat hreelayerconvolutionalarchitecture.Trainingandvalidationofthesuggestedmodelwereconductedonadatasetof1,200subjectsthatweredivided into ABO and Rh categories of the blood group.The experimental results show a mean classification of 93.4,sensitivity of 91.8 andspecificity of94.6 that were observed on the eight blood group classes.These findingsare much more accurate than PPG -based methodsthat hadbeen previously reported by about 6.2 percentage points.In addition, the use of majority voting on more than one PPG window also reduces themisclassificationsthat maybe causedby motionartifactsand signalnoise.The resulting systemissmall, low-cost, and highly applicabletopoint-ofcare implementation, which is a potentially valuable alternative to the traditional serological typing to perform a primary screening test inresource-constrained environments.

Fields of Study & Subjects

Non-Invasive Vital Sign Monitoring Biometric Identification and Security Blood groups and transfusion

Publication details

Journal / source
International journal of engineering science and advanced technology.
Volume
26
Issue
3
Pages
510–517
Publication date
1 March 2026
Language
EN