English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Monitoring medical data, e.g., Electrocardiogram (ECG) signals, is a commonapplication of Internet of Things (IoT) devices. Compression methods are oftenapplied on the massive amounts of sensor data generated prior to sending it tothe Cloud to reduce the storage and delivery costs. A lossy compressionprovides high compression gain (CG), but may reduce the performance of an ECGapplication (downstream task) due to information loss. Previous works on ECGmonitoring focus either on optimizing the signal reconstruction or the task'sperformance. Instead, we advocate a self-adapting lossy compression solutionthat enables configuring a desired performance level on the downstream taskswhile maintaining an optimized CG that reduces Cloud costs. We proposeDynamic-Deep, a task-aware compression geared for IoT-Cloud architectures. Ourcompressor is trained to optimize the CG while maintaining the performancerequirement of the downstream tasks chosen out of a wide range. In deployment,the IoT edge device adapts the compression and sends an optimizedrepresentation for each data segment, accounting for the downstream task'sdesired performance without relying on feedback from the Cloud. We conduct anextensive evaluation of our approach on common ECG datasets using two popularECG applications, which includes heart rate (HR) arrhythmia classification. Wedemonstrate that Dynamic-Deep can be configured to improve HR classificationF1-score in a wide range of requirements. One of which is tuned to improve theF1-score by 3 and increases CG by up to 83% compared to the previousstate-of-the-art (autoencoder-based) compressor. Analyzing Dynamic-Deep on theGoogle Cloud Platform, we observe a 97% reduction in cloud costs compared to ano compression solution.

Dynamic-Deep: Tune ECG Task Performance and Optimize Compression in IoT Architectures