Abstract
Deep neural networks (DNNs) have been successfully deployed in widespread domains, including healthcare applications. DenseNet201 is a new DNN architecture used in healthcare systems (i.e., presence detection of the surgical tool). Specialized accelerators such as GPUs have been used to speed up the execution of DNNs. Nevertheless, GPUs are prone to transient effects and other reliability threats, which can impact DNN models’ reliability. Safety-critical systems, such as healthcare applications, must be highly reliable because minor errors might lead to severe injury or death. In this paper, we propose a selective mitigation technique that relies on in-depth analysis. First, we inject the DenseNet201 model implemented on a GPU via NVIDIA’s SASSIFI fault injector. Second, we perform a comprehensive analysis from the perspective of kernel and layer to identify the most vulnerable portions of the injected model. Finally, we validate our technique by applying it to the top-vulnerable kernels to selectively protect the only sensitive portions of the model to avoid unnecessary overheads. Our experiments demonstrate that our mitigation technique achieves a significant reduction in the percentage of errors that cause malfunction (errors that lead to misclassification) from 6.463% to 0.21% . Moreover, the performance overhead (the execution time) of our technique is compared with the well-known protection techniques: Algorithm-Based Fault Tolerance (ABFT), Double Modular Redundancy (DMR), and Triple Modular Redundancy (TMR). The proposed solution shows only 0.3035% overhead compared to these techniques while correcting up 84.8% of the SDC errors in DenseNet201, remarkably improving the healthcare domain’s model reliability.
Highlights
Convolutional Neural Networks (CNNs) became prominent following the study conducted by LeCuN, where it was used to process grid-like data of images and time series [1]
As healthcare applications are safety-critical systems, they should be highly reliable because small errors might lead to serious injury or death [19]
We analyzed the DenseNet201 model trained on a dataset of presence detection of the surgical tool and evaluated the impact of Malfunction, Light-Malfunction, and No-Malfunction Silent Data Corruption (SDC) of the soft errors on the reliability of the DenseNet201 on Graphics Processing Units (GPUs)
Summary
Convolutional Neural Networks (CNNs) became prominent following the study conducted by LeCuN, where it was used to process grid-like data of images and time series [1]. CNNs are a special type of Deep Neural Networks (DNNs), which have shown state-of-the-art results on many competitive benchmarks. A. CONVOLUTIONAL NEURAL NETWORKS (CNNs) The CNN technique is the most commonly used approach for Deep Learning (DL). CONVOLUTIONAL NEURAL NETWORKS (CNNs) The CNN technique is the most commonly used approach for Deep Learning (DL) They have been more commonly used in healthcare applications such as scanning, diagnosing, generating reports, and treating various diseases [26]–[30]. The other two layers include the subsampling layer, which is optional, and the fully connected layers. The arrangement of these layers is in a feed-forward fashion [31]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
