Pubfacts - Scientific Publication Data
  • Categories
  • |
  • Journals
  • |
  • Authors
  • Login
  • Categories
  • Journals

Search Our Scientific Publications & Authors

Publications
  • Publications
  • Authors
find publications by category +
Translate page:

Real-time data analysis for medical diagnosis using FPGA-accelerated neural networks.

Authors:
Ahmed Sanaullah Chen Yang Yuri Alexeev Kazutomo Yoshii Martin C Herbordt

BMC Bioinformatics 2018 Dec 21;19(Suppl 18):490. Epub 2018 Dec 21.

Computer Architecture and Automated Design Lab, Boston University, Boston, MA, USA.

Background: Real-time analysis of patient data during medical procedures can provide vital diagnostic feedback that significantly improves chances of success. With sensors becoming increasingly fast, frameworks such as Deep Neural Networks are required to perform calculations within the strict timing constraints for real-time operation. However, traditional computing platforms responsible for running these algorithms incur a large overhead due to communication protocols, memory accesses, and static (often generic) architectures. In this work, we implement a low-latency Multi-Layer Perceptron (MLP) processor using Field Programmable Gate Arrays (FPGAs). Unlike CPUs and Graphics Processing Units (GPUs), our FPGA-based design can directly interface sensors, storage devices, display devices and even actuators, thus reducing the delays of data movement between ports and compute pipelines. Moreover, the compute pipelines themselves are tailored specifically to the application, improving resource utilization and reducing idle cycles. We demonstrate the effectiveness of our approach using mass-spectrometry data sets for real-time cancer detection.

Results: We demonstrate that correct parameter sizing, based on the application, can reduce latency by 20% on average. Furthermore, we show that in an application with tightly coupled data-path and latency constraints, having a large amount of computing resources can actually reduce performance. Using mass-spectrometry benchmarks, we show that our proposed FPGA design outperforms both CPU and GPU implementations, with an average speedup of 144x and 21x, respectively.

Conclusion: In our work, we demonstrate the importance of application-specific optimizations in order to minimize latency and maximize resource utilization for MLP inference. By directly interfacing and processing sensor data with ultra-low latency, FPGAs can perform real-time analysis during procedures and provide diagnostic feedback that can be critical to achieving higher percentages of successful patient outcomes.

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12859-018-2505-7DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6302367PMC
December 2018

Publication Analysis

Top Keywords

diagnostic feedback
8
resource utilization
8
compute pipelines
8
procedures provide
8
neural networks
8
real-time analysis
8
real-time
5
utilization reducing
4
reducing idle
4
idle cycles
4
cycles demonstrate
4
demonstrate effectiveness
4
effectiveness approach
4
approach mass-spectrometry
4
mass-spectrometry data
4
sets real-time
4
data sets
4
real-time cancer
4
cancer detectionresults
4
detectionresults demonstrate
4

Keyword Occurance

Similar Publications

Thyroid Feedback Quantile-based Index Is Associated with Blood Pressure and Other Hemodynamic Measures: A Cross-Sectional Study.

Authors:
Sijue Yang Zixiao Wang Jiarong Li Jinrong Fu Haixia Guan Wei Wang

Endocr Pract 2022 Aug 5. Epub 2022 Aug 5.

Department of Physical Examination Center, The First Hospital of China Medical University, Shenyang, P.R. China. Electronic address:

Objective: Previous researches reported inconsistent relationships between thyroid function and blood pressure. We aimed to further explore the associations between thyroid hormone sensitivity and blood pressure parameters.

Methods: This retrospective study included 6,272 subjects who underwent a health examination at the First Hospital of China Medical University between January 2017 and December 2018. Read More

View Article and Full-Text PDF
August 2022
Similar Publications

Research Progress on the Regulation Mechanism of Key Signal Pathways Affecting the Prognosis of Glioma.

Authors:
Hao Wu Min Wei Yuping Li Qiang Ma Hengzhu Zhang

Front Mol Neurosci 2022 22;15:910543. Epub 2022 Jul 22.

Graduate School of Dalian Medical University, Dalian, China.

As is known to all, glioma, a global difficult problem, has a high malignant degree, high recurrence rate and poor prognosis. We analyzed and summarized signal pathway of the Hippo/YAP, PI3K/AKT/mTOR, miRNA, WNT/β-catenin, Notch, Hedgehog, TGF-β, TCS/mTORC1 signal pathway, JAK/STAT signal pathway, MAPK signaling pathway, the relationship between BBB and signal pathways and the mechanism of key enzymes in glioma. It is concluded that Yap1 inhibitor may become an effective target for the treatment of glioma in the near future through efforts of generation after generation. Read More

View Article and Full-Text PDF
July 2022
Similar Publications

Routine data-based quality indicators for the treatment of gonarthrosis and coxarthrosis patients in the ambulatory care sector - A study protocol for a cluster-randomised pilot trial to evaluate the MobilE-ARTH study.

Authors:
Tobias Bock Ronja Flemming Wiebke Schüttig Anja Schramm Martin B Weigl Leonie Sundmacher

BMC Musculoskelet Disord 2022 Aug 4;23(1):747. Epub 2022 Aug 4.

Chair of Health Economics, Department of Sport and Health Sciences, Technical University of Munich (TUM), Munich, Germany.

Background: In 2019, Germany had the highest rate of hip replacement surgery and the fourth highest rate of knee replacement surgery among more than 30 OECD countries. The age-standardised rates were estimated at 174 hip joint and 137 knee joint replacements per 100,000 population. Against this background, the contrast between financial incentives for surgery and missing incentives for non-surgical treatment options is repeatedly discussed. Read More

View Article and Full-Text PDF
August 2022
Similar Publications

A Systematic Review of Interventions to Reduce Computed Tomography Usage in the Emergency Department.

Authors:
Cody L Dunne Jason L Elzinga Allen Vorobeichik Sanjana Sudershan Diana Keto-Lambert Eddy Lang Shawn Dowling

Ann Emerg Med 2022 Aug 1. Epub 2022 Aug 1.

Department of Emergency Medicine, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada.

Study Objective: Unnecessary computed tomography (CT) scans burden the health care system, leading to increased emergency department (ED) wait times and lengths of stay, costing almost a billion dollars annually. This study aimed to describe ED-based interventions that are most effective at reducing CT imaging while maintaining diagnostic accuracy and patient safety.

Methods: Adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, MEDLINE, Embase, CINAHL, Cochrane Central Register of Controlled Trials, and Google Scholar were searched until December 31, 2020. Read More

View Article and Full-Text PDF
August 2022
Similar Publications

[Chinese expert consensus on the management of lower respiratory tract infections of in adults(2022)].

Authors:

Zhonghua Jie He He Hu Xi Za Zhi 2022 Aug;45(8):739-752

(PA) is the second common Gram-negative bacterium for hospital acquired pneumonia (HAP) in China (16.9%-22.0%). Read More

View Article and Full-Text PDF
August 2022
Similar Publications
}
© 2022 PubFacts.
  • About PubFacts
  • Privacy Policy
  • Sitemap