A 1.2-0.55V general-purpose biomedical processor with configurable machine-learning accelerators for high-order, patient-adaptive monitoring

Kyong Ho Lee; Naveen Verma

doi:10.1109/ESSCIRC.2012.6341275

A 1.2-0.55V general-purpose biomedical processor with configurable machine-learning accelerators for high-order, patient-adaptive monitoring

Kyong Ho Lee, Naveen Verma

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

13 Scopus citations

Abstract

Machine learning offers powerful advantages in sensing systems, enabling the creation and adaptation of highorder signal models by exploiting the sensed data. We present a general-purpose processor that employs configurable machine-learning accelerators to analyze physiological signals at low energy levels for a broad range of biomedical applications. Implemented in 130nm LP CMOS, the processor operates from 1.2V-0.55V (logic). It achieves real-time EEG-based seizure detection at 273μW (at 0.85V) and patient-adaptive ECG-based cardiac-arrhythmia detection at 124μW (at 0.75V), yielding overall energy savings of 62.4× and 144.7× thanks to the accelerators.

Original language	English (US)
Title of host publication	2012 Proceedings of the European Solid State Circuits Conference, ESSCIRC 2012
Pages	285-288
Number of pages	4
DOIs	https://doi.org/10.1109/ESSCIRC.2012.6341275
State	Published - Dec 14 2012
Event	38th European Solid State Circuits Conference, ESSCIRC 2012 - Bordeaux, France Duration: Sep 17 2012 → Sep 21 2012

Publication series

Name	European Solid-State Circuits Conference
ISSN (Print)	1930-8833

Other

Other	38th European Solid State Circuits Conference, ESSCIRC 2012
Country/Territory	France
City	Bordeaux
Period	9/17/12 → 9/21/12

All Science Journal Classification (ASJC) codes

Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/ESSCIRC.2012.6341275

Cite this

@inproceedings{bb1d1662d8894bb9b6652284f44aa807,

title = "A 1.2-0.55V general-purpose biomedical processor with configurable machine-learning accelerators for high-order, patient-adaptive monitoring",

abstract = "Machine learning offers powerful advantages in sensing systems, enabling the creation and adaptation of highorder signal models by exploiting the sensed data. We present a general-purpose processor that employs configurable machine-learning accelerators to analyze physiological signals at low energy levels for a broad range of biomedical applications. Implemented in 130nm LP CMOS, the processor operates from 1.2V-0.55V (logic). It achieves real-time EEG-based seizure detection at 273μW (at 0.85V) and patient-adaptive ECG-based cardiac-arrhythmia detection at 124μW (at 0.75V), yielding overall energy savings of 62.4× and 144.7× thanks to the accelerators.",

author = "Lee, {Kyong Ho} and Naveen Verma",

year = "2012",

month = dec,

day = "14",

doi = "10.1109/ESSCIRC.2012.6341275",

language = "English (US)",

isbn = "9781467322126",

series = "European Solid-State Circuits Conference",

pages = "285--288",

booktitle = "2012 Proceedings of the European Solid State Circuits Conference, ESSCIRC 2012",

note = "38th European Solid State Circuits Conference, ESSCIRC 2012 ; Conference date: 17-09-2012 Through 21-09-2012",

}

Lee, KH & Verma, N 2012, A 1.2-0.55V general-purpose biomedical processor with configurable machine-learning accelerators for high-order, patient-adaptive monitoring. in 2012 Proceedings of the European Solid State Circuits Conference, ESSCIRC 2012., 6341275, European Solid-State Circuits Conference, pp. 285-288, 38th European Solid State Circuits Conference, ESSCIRC 2012, Bordeaux, France, 9/17/12. https://doi.org/10.1109/ESSCIRC.2012.6341275

A 1.2-0.55V general-purpose biomedical processor with configurable machine-learning accelerators for high-order, patient-adaptive monitoring. / Lee, Kyong Ho; Verma, Naveen.
2012 Proceedings of the European Solid State Circuits Conference, ESSCIRC 2012. 2012. p. 285-288 6341275 (European Solid-State Circuits Conference).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - A 1.2-0.55V general-purpose biomedical processor with configurable machine-learning accelerators for high-order, patient-adaptive monitoring

AU - Lee, Kyong Ho

AU - Verma, Naveen

PY - 2012/12/14

Y1 - 2012/12/14

N2 - Machine learning offers powerful advantages in sensing systems, enabling the creation and adaptation of highorder signal models by exploiting the sensed data. We present a general-purpose processor that employs configurable machine-learning accelerators to analyze physiological signals at low energy levels for a broad range of biomedical applications. Implemented in 130nm LP CMOS, the processor operates from 1.2V-0.55V (logic). It achieves real-time EEG-based seizure detection at 273μW (at 0.85V) and patient-adaptive ECG-based cardiac-arrhythmia detection at 124μW (at 0.75V), yielding overall energy savings of 62.4× and 144.7× thanks to the accelerators.

AB - Machine learning offers powerful advantages in sensing systems, enabling the creation and adaptation of highorder signal models by exploiting the sensed data. We present a general-purpose processor that employs configurable machine-learning accelerators to analyze physiological signals at low energy levels for a broad range of biomedical applications. Implemented in 130nm LP CMOS, the processor operates from 1.2V-0.55V (logic). It achieves real-time EEG-based seizure detection at 273μW (at 0.85V) and patient-adaptive ECG-based cardiac-arrhythmia detection at 124μW (at 0.75V), yielding overall energy savings of 62.4× and 144.7× thanks to the accelerators.

UR - http://www.scopus.com/inward/record.url?scp=84870801482&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870801482&partnerID=8YFLogxK

U2 - 10.1109/ESSCIRC.2012.6341275

DO - 10.1109/ESSCIRC.2012.6341275

M3 - Conference contribution

AN - SCOPUS:84870801482

SN - 9781467322126

T3 - European Solid-State Circuits Conference

SP - 285

EP - 288

BT - 2012 Proceedings of the European Solid State Circuits Conference, ESSCIRC 2012

T2 - 38th European Solid State Circuits Conference, ESSCIRC 2012

Y2 - 17 September 2012 through 21 September 2012

ER -

A 1.2-0.55V general-purpose biomedical processor with configurable machine-learning accelerators for high-order, patient-adaptive monitoring

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this