Key result
CaliBench automates benchmarking to standardize and evaluate computational methods for ANS data analysis.
Why the study?
Division of labour between empirical data acquisition and AI data analytics is rising, creating a gap between experimental laboratories and analytical research groups.
This methodology proposal introduces an automated benchmarking framework to bridge the gap between experimental neuroscience data acquisition and the development of computational data analytics.
In many fields of science, division of labour between acquisition of empirical data and development of data analytics is on the rise, and a recent emphasis of artificial intelligence-based data analytics has a potential to accelerate this trend. A case in point is the use of autonomic nervous system readouts in neuroscience and psychology. These are often generated by laboratories with strong experimental focus, while data analytics are increasingly developed in research groups with focus in bioengineering, numerical optimization, or machine learning. Here, we propose a strategy to bridge this gap with three pillars: (a) a common data and methods standard, (b) dynamic curated benchmark data sets, and (c) an automated benchmarking engine. We describe these components and propose details for their implementation.
No takes yet. Share an insight, caveat, or question.
Kulkarni et al. (2023) studied this question. CaliBench was evaluated. The authors propose CaliBench, an automated calibration benchmarking engine to standardize and evaluate computational methods for analyzing autonomic nervous system data.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: