In-Sensor Artificial Intelligence and Fusion With Electronic Medical Records for At-Home Monitoring

Sudarsan Sadasivuni; Monjoy Saha; Sumukh Prashant Bhanushali; Imon Banerjee; Arindam Sanyal

doi:10.1109/TBCAS.2023.3251310

In-Sensor Artificial Intelligence and Fusion With Electronic Medical Records for At-Home Monitoring

Sudarsan Sadasivuni, Monjoy Saha, Sumukh Prashant Bhanushali, Imon Banerjee, Arindam Sanyal

Diagnostic Radiology

Research output: Contribution to journal › Article › peer-review

Abstract

This work presents an artificial intelligence (AI) framework for real-time, personalized sepsis prediction four hours before onset through fusion of electrocardiogram (ECG) and patient electronic medical record. An on-chip classifier combines analog reservoir-computer and artificial neural network to perform prediction without front-end data converter or feature extraction which reduces energy by 13× compared to digital baseline at normalized power efficiency of 528 TOPS/W, and reduces energy by 159× compared to RF transmission of all digitized ECG samples. The proposed AI framework predicts sepsis onset with 89.9% and 92.9% accuracy on patient data from Emory University Hospital and MIMIC-III respectively. The proposed framework is non-invasive and does not require lab tests which makes it suitable for at-home monitoring.

Original language	English (US)
Pages (from-to)	312-322
Number of pages	11
Journal	IEEE Transactions on Biomedical Circuits and Systems
Volume	17
Issue number	2
DOIs	https://doi.org/10.1109/TBCAS.2023.3251310
State	Published - Apr 1 2023

Keywords

Sepsis
artificial intelligence
artificial neural network
data fusion
in-memory computing
reservoir-computer

ASJC Scopus subject areas

Biomedical Engineering
Electrical and Electronic Engineering

Access to Document

10.1109/TBCAS.2023.3251310

Cite this

@article{c4d9b5b445b4443488a8cb550987b1ee,

title = "In-Sensor Artificial Intelligence and Fusion With Electronic Medical Records for At-Home Monitoring",

abstract = "This work presents an artificial intelligence (AI) framework for real-time, personalized sepsis prediction four hours before onset through fusion of electrocardiogram (ECG) and patient electronic medical record. An on-chip classifier combines analog reservoir-computer and artificial neural network to perform prediction without front-end data converter or feature extraction which reduces energy by 13× compared to digital baseline at normalized power efficiency of 528 TOPS/W, and reduces energy by 159× compared to RF transmission of all digitized ECG samples. The proposed AI framework predicts sepsis onset with 89.9% and 92.9% accuracy on patient data from Emory University Hospital and MIMIC-III respectively. The proposed framework is non-invasive and does not require lab tests which makes it suitable for at-home monitoring.",

keywords = "Sepsis, artificial intelligence, artificial neural network, data fusion, in-memory computing, reservoir-computer",

author = "Sudarsan Sadasivuni and Monjoy Saha and Bhanushali, {Sumukh Prashant} and Imon Banerjee and Arindam Sanyal",

note = "Publisher Copyright: {\textcopyright} 2007-2012 IEEE.",

year = "2023",

month = apr,

day = "1",

doi = "10.1109/TBCAS.2023.3251310",

language = "English (US)",

volume = "17",

pages = "312--322",

journal = "IEEE Transactions on Biomedical Circuits and Systems",

issn = "1932-4545",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2",

}

TY - JOUR

T1 - In-Sensor Artificial Intelligence and Fusion With Electronic Medical Records for At-Home Monitoring

AU - Sadasivuni, Sudarsan

AU - Saha, Monjoy

AU - Bhanushali, Sumukh Prashant

AU - Banerjee, Imon

AU - Sanyal, Arindam

PY - 2023/4/1

Y1 - 2023/4/1

N2 - This work presents an artificial intelligence (AI) framework for real-time, personalized sepsis prediction four hours before onset through fusion of electrocardiogram (ECG) and patient electronic medical record. An on-chip classifier combines analog reservoir-computer and artificial neural network to perform prediction without front-end data converter or feature extraction which reduces energy by 13× compared to digital baseline at normalized power efficiency of 528 TOPS/W, and reduces energy by 159× compared to RF transmission of all digitized ECG samples. The proposed AI framework predicts sepsis onset with 89.9% and 92.9% accuracy on patient data from Emory University Hospital and MIMIC-III respectively. The proposed framework is non-invasive and does not require lab tests which makes it suitable for at-home monitoring.

AB - This work presents an artificial intelligence (AI) framework for real-time, personalized sepsis prediction four hours before onset through fusion of electrocardiogram (ECG) and patient electronic medical record. An on-chip classifier combines analog reservoir-computer and artificial neural network to perform prediction without front-end data converter or feature extraction which reduces energy by 13× compared to digital baseline at normalized power efficiency of 528 TOPS/W, and reduces energy by 159× compared to RF transmission of all digitized ECG samples. The proposed AI framework predicts sepsis onset with 89.9% and 92.9% accuracy on patient data from Emory University Hospital and MIMIC-III respectively. The proposed framework is non-invasive and does not require lab tests which makes it suitable for at-home monitoring.

KW - Sepsis

KW - artificial intelligence

KW - artificial neural network

KW - data fusion

KW - in-memory computing

KW - reservoir-computer

UR - http://www.scopus.com/inward/record.url?scp=85149464024&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85149464024&partnerID=8YFLogxK

U2 - 10.1109/TBCAS.2023.3251310

DO - 10.1109/TBCAS.2023.3251310

M3 - Article

C2 - 37028013

AN - SCOPUS:85149464024

SN - 1932-4545

VL - 17

SP - 312

EP - 322

JO - IEEE Transactions on Biomedical Circuits and Systems

JF - IEEE Transactions on Biomedical Circuits and Systems

IS - 2

ER -

In-Sensor Artificial Intelligence and Fusion With Electronic Medical Records for At-Home Monitoring

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this