Hardware-oriented pruning and quantization of Deep Learning models to detect life-threatening arrhythmias

Gonzalez-Carabarin, LizethSchmid, AlexandreVan Sloun, Ruud J. G.2022-08-292022-08-292022-08-292021-01-0110.1109/BIOCAS49922.2021.9644976https://infoscience.epfl.ch/handle/20.500.14299/190326WOS:000837980700049Wearable solutions based on Deep Learning (DL) for real-time ECG monitoring are a promising alternative to detect life-threatening arrhythmias. However, DL models suffer of a large memory footprint, which hampers their adoption in portable technologies. Therefore, we leverage a hardware-oriented pruning approach to effectively shrink DL models. We demonstrate that tiny DL models can be reduced to 5.55x (pruning), and 26.6x (pruning+quantization) compression rate, with 82.9% FLOP's reduction. These ultra-compressed models are able to effectively classify life-threatening arrhythmias with minimal or no loss of performance compared with their non=pruned counterparts, which can pave the path towards DL-based biomedical portable solutions.Computer Science, Information SystemsComputer Science, Interdisciplinary ApplicationsEngineering, BiomedicalEngineering, Electrical & ElectronicComputer ScienceEngineeringecgshockable arrhythmiadeep learningpruningmodel compressionquantizationHardware-oriented pruning and quantization of Deep Learning models to detect life-threatening arrhythmiastext::conference output::conference proceedings::conference paper