e-G2C: A 0.14-to-8.31 J/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM

Yang Zhao; Yongan Zhang; Yonggan Fu; Xu Ouyang; Cheng Wan; Shang Wu; Anton Banta; Mathews M. John; Allison Post; Mehdi Razavi; Joseph Cavallaro; Behnaam Aazhang; Yingyan Lin

doi:10.1109/VLSITechnologyandCir46769.2022.9830335

e-G2C: A 0.14-to-8.31 J/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM

Yang Zhao
, Yongan Zhang
, Yonggan Fu
, Xu Ouyang
, Cheng Wan
, Shang Wu
, Anton Banta
, Mathews M. John
, Allison Post
, Mehdi Razavi
, Joseph Cavallaro
, Behnaam Aazhang
, Yingyan Lin

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

4 Scopus citations

Abstract

This work presents the first silicon-validated dedicated EGM-to-ECG (G2C) processor, dubbed e-G2C, featuring continuous lightweight anomaly detection, event-driven coarse/precise conversion, and on-chip adaptation. e-G2C utilizes neural network (NN) based G2C conversion and integrates 1) an architecture supporting anomaly detection and coarse/precise conversion via time multiplexing to balance the effectiveness and power, 2) an algorithm-hardware co-designed vector-wise sparsity resulting in a 1.6-1.7× speedup, 3) hybrid dataflows for enhancing near 100% utilization for normal/depth-wise(DW)/point-wise(PW) convolutions (Convs), and 4) an on-chip detection threshold adaptation engine for continuous effectiveness. The achieved 0.14-8.31 J/inference energy efficiency outperforms prior arts under similar complexity, promising real-time detection/conversion and possibly life-critical interventions.

Original language	English (US)
Title of host publication	2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	252-253
Number of pages	2
ISBN (Electronic)	9781665497725
DOIs	https://doi.org/10.1109/VLSITechnologyandCir46769.2022.9830335
State	Published - 2022
Externally published	Yes
Event	2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022 - Honolulu, United States Duration: Jun 12 2022 → Jun 17 2022

Publication series

Name	Digest of Technical Papers - Symposium on VLSI Technology
Volume	2022-June
ISSN (Print)	0743-1562

Conference

Conference	2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022
Country/Territory	United States
City	Honolulu
Period	6/12/22 → 6/17/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Publisher link

10.1109/VLSITechnologyandCir46769.2022.9830335

Find It

Find It at U of M Libraries

Cite this

Zhao, Y., Zhang, Y., Fu, Y., Ouyang, X., Wan, C., Wu, S., Banta, A., John, M. M., Post, A., Razavi, M., Cavallaro, J., Aazhang, B., & Lin, Y. (2022). e-G2C: A 0.14-to-8.31 J/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM. In 2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022 (pp. 252-253). (Digest of Technical Papers - Symposium on VLSI Technology; Vol. 2022-June). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/VLSITechnologyandCir46769.2022.9830335

e-G2C: A 0.14-to-8.31 J/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM. / Zhao, Yang; Zhang, Yongan; Fu, Yonggan et al.
2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022. Institute of Electrical and Electronics Engineers Inc., 2022. p. 252-253 (Digest of Technical Papers - Symposium on VLSI Technology; Vol. 2022-June).

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

Zhao, Y, Zhang, Y, Fu, Y, Ouyang, X, Wan, C, Wu, S, Banta, A, John, MM, Post, A, Razavi, M, Cavallaro, J, Aazhang, B & Lin, Y 2022, e-G2C: A 0.14-to-8.31 J/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM. in 2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022. Digest of Technical Papers - Symposium on VLSI Technology, vol. 2022-June, Institute of Electrical and Electronics Engineers Inc., pp. 252-253, 2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022, Honolulu, United States, 6/12/22. https://doi.org/10.1109/VLSITechnologyandCir46769.2022.9830335

Zhao Y, Zhang Y, Fu Y, Ouyang X, Wan C, Wu S et al. e-G2C: A 0.14-to-8.31 J/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM. In 2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 252-253. (Digest of Technical Papers - Symposium on VLSI Technology). doi: 10.1109/VLSITechnologyandCir46769.2022.9830335

Zhao, Yang ; Zhang, Yongan ; Fu, Yonggan et al. / e-G2C : A 0.14-to-8.31 J/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM. 2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 252-253 (Digest of Technical Papers - Symposium on VLSI Technology).

@inproceedings{2153bd35a56d416085aeed74737771c6,

title = "e-G2C: A 0.14-to-8.31 J/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM",

abstract = "This work presents the first silicon-validated dedicated EGM-to-ECG (G2C) processor, dubbed e-G2C, featuring continuous lightweight anomaly detection, event-driven coarse/precise conversion, and on-chip adaptation. e-G2C utilizes neural network (NN) based G2C conversion and integrates 1) an architecture supporting anomaly detection and coarse/precise conversion via time multiplexing to balance the effectiveness and power, 2) an algorithm-hardware co-designed vector-wise sparsity resulting in a 1.6-1.7× speedup, 3) hybrid dataflows for enhancing near 100\% utilization for normal/depth-wise(DW)/point-wise(PW) convolutions (Convs), and 4) an on-chip detection threshold adaptation engine for continuous effectiveness. The achieved 0.14-8.31 J/inference energy efficiency outperforms prior arts under similar complexity, promising real-time detection/conversion and possibly life-critical interventions.",

author = "Yang Zhao and Yongan Zhang and Yonggan Fu and Xu Ouyang and Cheng Wan and Shang Wu and Anton Banta and John, \{Mathews M.\} and Allison Post and Mehdi Razavi and Joseph Cavallaro and Behnaam Aazhang and Yingyan Lin",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022 ; Conference date: 12-06-2022 Through 17-06-2022",

year = "2022",

doi = "10.1109/VLSITechnologyandCir46769.2022.9830335",

language = "English (US)",

series = "Digest of Technical Papers - Symposium on VLSI Technology",

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

pages = "252--253",

booktitle = "2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022",

}

TY - GEN

T1 - e-G2C

T2 - 2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022

AU - Zhao, Yang

AU - Zhang, Yongan

AU - Fu, Yonggan

AU - Ouyang, Xu

AU - Wan, Cheng

AU - Wu, Shang

AU - Banta, Anton

AU - John, Mathews M.

AU - Post, Allison

AU - Razavi, Mehdi

AU - Cavallaro, Joseph

AU - Aazhang, Behnaam

AU - Lin, Yingyan

PY - 2022

Y1 - 2022

N2 - This work presents the first silicon-validated dedicated EGM-to-ECG (G2C) processor, dubbed e-G2C, featuring continuous lightweight anomaly detection, event-driven coarse/precise conversion, and on-chip adaptation. e-G2C utilizes neural network (NN) based G2C conversion and integrates 1) an architecture supporting anomaly detection and coarse/precise conversion via time multiplexing to balance the effectiveness and power, 2) an algorithm-hardware co-designed vector-wise sparsity resulting in a 1.6-1.7× speedup, 3) hybrid dataflows for enhancing near 100% utilization for normal/depth-wise(DW)/point-wise(PW) convolutions (Convs), and 4) an on-chip detection threshold adaptation engine for continuous effectiveness. The achieved 0.14-8.31 J/inference energy efficiency outperforms prior arts under similar complexity, promising real-time detection/conversion and possibly life-critical interventions.

AB - This work presents the first silicon-validated dedicated EGM-to-ECG (G2C) processor, dubbed e-G2C, featuring continuous lightweight anomaly detection, event-driven coarse/precise conversion, and on-chip adaptation. e-G2C utilizes neural network (NN) based G2C conversion and integrates 1) an architecture supporting anomaly detection and coarse/precise conversion via time multiplexing to balance the effectiveness and power, 2) an algorithm-hardware co-designed vector-wise sparsity resulting in a 1.6-1.7× speedup, 3) hybrid dataflows for enhancing near 100% utilization for normal/depth-wise(DW)/point-wise(PW) convolutions (Convs), and 4) an on-chip detection threshold adaptation engine for continuous effectiveness. The achieved 0.14-8.31 J/inference energy efficiency outperforms prior arts under similar complexity, promising real-time detection/conversion and possibly life-critical interventions.

UR - https://www.scopus.com/pages/publications/85135221777

UR - https://www.scopus.com/pages/publications/85135221777#tab=citedBy

U2 - 10.1109/VLSITechnologyandCir46769.2022.9830335

DO - 10.1109/VLSITechnologyandCir46769.2022.9830335

M3 - Conference contribution

AN - SCOPUS:85135221777

T3 - Digest of Technical Papers - Symposium on VLSI Technology

SP - 252

EP - 253

BT - 2022 IEEE Symposium on VLSI Technology and Circuits, VLSI Technology and Circuits 2022

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 12 June 2022 through 17 June 2022

ER -

e-G2C: A 0.14-to-8.31 J/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM

Abstract

Publication series

Conference

Bibliographical note

UN SDGs

Publisher link

Find It

Other files and links

Fingerprint

Cite this