Home > Research > Publications & Outputs > A high-performance, hardware-based deep learnin...

Links

Text available via DOI:

View graph of relations

A high-performance, hardware-based deep learning system for disease diagnosis

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

A high-performance, hardware-based deep learning system for disease diagnosis. / Siddique, Ali; Iqbal, Muhammad Azhar; Aleem, Muhammad et al.

In: PeerJ Computer Science, Vol. 8, e1034, 19.07.2022.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Siddique, A., Iqbal, M. A., Aleem, M., & Lin, J. C-W. (2022). A high-performance, hardware-based deep learning system for disease diagnosis. PeerJ Computer Science, 8, [e1034]. https://doi.org/10.7717/peerj-cs.1034

Vancouver

Siddique A, Iqbal MA, Aleem M, Lin JC-W. A high-performance, hardware-based deep learning system for disease diagnosis. PeerJ Computer Science. 2022 Jul 19;8:e1034. doi: 10.7717/peerj-cs.1034

Author

Siddique, Ali ; Iqbal, Muhammad Azhar ; Aleem, Muhammad et al. / A high-performance, hardware-based deep learning system for disease diagnosis. In: PeerJ Computer Science. 2022 ; Vol. 8.

Bibtex

@article{b75692db281e4f559ec7dc011991be50,
title = "A high-performance, hardware-based deep learning system for disease diagnosis",
abstract = "Modern deep learning schemes have shown human-level performance in the area of medical science. However, the implementation of deep learning algorithms on dedicated hardware remains a challenging task because modern algorithms and neuronal activation functions are generally not hardware-friendly and require a lot of resources. Recently, researchers have come up with some hardware-friendly activation functions that can yield high throughput and high accuracy at the same time. In this context, we propose a hardware-based neural network that can predict the presence of cancer in humans with 98.23% accuracy. This is done by making use of cost-efficient, highly accurate activation functions, Sqish and LogSQNL. Due to its inherently parallel components, the system can classify a given sample in just one clock cycle, i.e., 15.75 nanoseconds. Though this system is dedicated to cancer diagnosis, it can predict the presence of many other diseases such as those of the heart. This is because the system is reconfigurable and can be programmed to classify any sample into one of two classes. The proposed hardware system requires about 983 slice registers, 2,655 slice look-up tables, and only 1.1 kilobits of on-chip memory. The system can predict about 63.5 million cancer samples in a second and can perform about 20 giga-operations per second. The proposed system is about 5–16 times cheaper and at least four times speedier than other dedicated hardware systems using neural networks for classification tasks.",
keywords = "Activation function, Cancer diagnosis, Deep learning, Field programmable gate array, Hardware friendly, Neural networks, Swish",
author = "Ali Siddique and Iqbal, {Muhammad Azhar} and Muhammad Aleem and Lin, {Jerry Chun-Wei}",
year = "2022",
month = jul,
day = "19",
doi = "10.7717/peerj-cs.1034",
language = "English",
volume = "8",
journal = "PeerJ Computer Science",
issn = "2376-5992",
publisher = "PeerJ Inc.",

}

RIS

TY - JOUR

T1 - A high-performance, hardware-based deep learning system for disease diagnosis

AU - Siddique, Ali

AU - Iqbal, Muhammad Azhar

AU - Aleem, Muhammad

AU - Lin, Jerry Chun-Wei

PY - 2022/7/19

Y1 - 2022/7/19

N2 - Modern deep learning schemes have shown human-level performance in the area of medical science. However, the implementation of deep learning algorithms on dedicated hardware remains a challenging task because modern algorithms and neuronal activation functions are generally not hardware-friendly and require a lot of resources. Recently, researchers have come up with some hardware-friendly activation functions that can yield high throughput and high accuracy at the same time. In this context, we propose a hardware-based neural network that can predict the presence of cancer in humans with 98.23% accuracy. This is done by making use of cost-efficient, highly accurate activation functions, Sqish and LogSQNL. Due to its inherently parallel components, the system can classify a given sample in just one clock cycle, i.e., 15.75 nanoseconds. Though this system is dedicated to cancer diagnosis, it can predict the presence of many other diseases such as those of the heart. This is because the system is reconfigurable and can be programmed to classify any sample into one of two classes. The proposed hardware system requires about 983 slice registers, 2,655 slice look-up tables, and only 1.1 kilobits of on-chip memory. The system can predict about 63.5 million cancer samples in a second and can perform about 20 giga-operations per second. The proposed system is about 5–16 times cheaper and at least four times speedier than other dedicated hardware systems using neural networks for classification tasks.

AB - Modern deep learning schemes have shown human-level performance in the area of medical science. However, the implementation of deep learning algorithms on dedicated hardware remains a challenging task because modern algorithms and neuronal activation functions are generally not hardware-friendly and require a lot of resources. Recently, researchers have come up with some hardware-friendly activation functions that can yield high throughput and high accuracy at the same time. In this context, we propose a hardware-based neural network that can predict the presence of cancer in humans with 98.23% accuracy. This is done by making use of cost-efficient, highly accurate activation functions, Sqish and LogSQNL. Due to its inherently parallel components, the system can classify a given sample in just one clock cycle, i.e., 15.75 nanoseconds. Though this system is dedicated to cancer diagnosis, it can predict the presence of many other diseases such as those of the heart. This is because the system is reconfigurable and can be programmed to classify any sample into one of two classes. The proposed hardware system requires about 983 slice registers, 2,655 slice look-up tables, and only 1.1 kilobits of on-chip memory. The system can predict about 63.5 million cancer samples in a second and can perform about 20 giga-operations per second. The proposed system is about 5–16 times cheaper and at least four times speedier than other dedicated hardware systems using neural networks for classification tasks.

KW - Activation function

KW - Cancer diagnosis

KW - Deep learning

KW - Field programmable gate array

KW - Hardware friendly

KW - Neural networks

KW - Swish

U2 - 10.7717/peerj-cs.1034

DO - 10.7717/peerj-cs.1034

M3 - Journal article

VL - 8

JO - PeerJ Computer Science

JF - PeerJ Computer Science

SN - 2376-5992

M1 - e1034

ER -