A Time-Frequency Deep Learning Classification Model for Metal Oxide Coated Particles

TY - GEN

T1 - A Time-Frequency Deep Learning Classification Model for Metal Oxide Coated Particles

AU - Tahir, Muhammad Nabeel

AU - Ashley, Brandon K.

AU - Sui, Jianye

AU - Javanmard, Mehdi

AU - Hassan, Umer

N1 - Funding Information: The authors would like to acknowledge the funding support from National Science Foundation (Award # 2002511), NIH’s training grant (T32 GM135141) and VCRI Exploratory Research Seed Grant by Rutgers. The authors also acknowledge the funding support from School of Engineering, and Global Health Institure at Rutgers University. Publisher Copyright: © 2023 IEEE.

PY - 2023

Y1 - 2023

N2 - This study uses time-frequency transformed data and deep learning (DL) models to identify the groups of metal oxide nano-coated micro-particles using an impedance cytometer. The nano-coated bioparticles generate distinct electrical signals in a multifrequency electric field and can be used in biosensing applications. The current machine learning-enabled sensing modalities are unable to accurately differentiate different bioparticles as the feature selection and feature engineering techniques are ineffective in selecting useful and informative features. Here, we use Wigner-Vile Distribution to transform the time series data into the time-frequency domain and employ three deep learning models to evaluate the ability of time-frequency transformed data to accurately represent the most important features. A classification accuracy of 75% for (10nm and 30nm) coated particles was achieved on the simplest DL model. This combination of time-frequency representation and the DL model will be sufficient to differentiate bioparticles by acting as an alternative to other ML-based techniques.

AB - This study uses time-frequency transformed data and deep learning (DL) models to identify the groups of metal oxide nano-coated micro-particles using an impedance cytometer. The nano-coated bioparticles generate distinct electrical signals in a multifrequency electric field and can be used in biosensing applications. The current machine learning-enabled sensing modalities are unable to accurately differentiate different bioparticles as the feature selection and feature engineering techniques are ineffective in selecting useful and informative features. Here, we use Wigner-Vile Distribution to transform the time series data into the time-frequency domain and employ three deep learning models to evaluate the ability of time-frequency transformed data to accurately represent the most important features. A classification accuracy of 75% for (10nm and 30nm) coated particles was achieved on the simplest DL model. This combination of time-frequency representation and the DL model will be sufficient to differentiate bioparticles by acting as an alternative to other ML-based techniques.

KW - Deep Learning

KW - impedance cytometer

KW - microfluidic

KW - multiplexing

UR - http://www.scopus.com/inward/record.url?scp=85165652406&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85165652406&partnerID=8YFLogxK

U2 - https://doi.org/10.1109/MDTS58049.2023.10168045

DO - https://doi.org/10.1109/MDTS58049.2023.10168045

M3 - Conference contribution

T3 - 2023 IEEE Microelectronics Design and Test Symposium, MDTS 2023

BT - 2023 IEEE Microelectronics Design and Test Symposium, MDTS 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 32nd IEEE Microelectronics Design and Test Symposium, MDTS 2023

Y2 - 8 May 2023 through 10 May 2023

ER -