Abstract
The rapid proliferation of wearable devices for medical applications has necessitated the need for automated algorithms to provide labelling of physiological time‐series data to identify abnormal morphology. However, such algorithms are less reliable than gold‐standard human expert labels (where the latter are typically difficult and expensive to obtain), due to their large inter‐ and intra‐subject variabilities. Actions taken in response to these algorithms can therefore result in sub‐optimal patient care. In a typical scenario where only unevenly sampled continuous or numeric estimates are provided, without access to the “ground truth”, it is challenging to choose which algorithms to trust and which to ignore, or even how to merge the outputs from multiple algorithms to form a more precise final estimate for individual patients. In this work, the novel application of two previously proposed parametric fully‐Bayesian graphical models is demonstrated for fusing labels from (i) independent and (ii) potentially correlated algorithms, validated on two publicly available datasets for the task of respiratory rate (RR) estimation. These unsupervised models aggregate RR labels and estimate jointly the assumed bias and precision of each algorithm. Fusing estimates in this way may then be used to infer the underlying ground truth for individual patients. It is shown that modelling the latent correlations between algorithms improves the RR estimates, when compared to commonly employed strategies in the literature. Finally, it is demonstrated that the adoption of a strongly Bayesian approach to inference using Gibbs sampling results in improved estimation over the current state‐of‐the‐art (e.g. hierarchical Gaussian processes) in physiological time‐series modelling.
Highlights
AND RELATED WORKWith the rapid increase in the volume and variety of wearable devices routinely in use for healthcare applications, there exists the possibility of personalising the care patients receive based on their individual physiologies
To date, automated algorithms remain less reliable in practice than labelling from human experts
We demonstrate improved estimation of respiratory rate (RR) is possible using our approach of fusing labels from different annotators, when compared with existing methods presented in the literature; namely two expectation maximisation (EM) models by [1] and [2], as well as a hierarchical Gaussian process approach [8]
Summary
With the rapid increase in the volume and variety of wearable devices routinely in use for healthcare applications, there exists the possibility of personalising the care patients receive based on their individual physiologies. A Bayesian EM framework which fused binary, multi-valued and continuous-valued labels was proposed in [4] This method described explicitly modelling the precision (but not bias) of individual annotators by taking into account their different skill levels. We propose to compare the performance of both parametric and non-parametric models and determine experimentally which method is more suitable for modelling physiological time-series data in the case when combining multiple imperfect algorithms to form a consensus, using the two public datasets as exemplars. We demonstrate improved estimation of RR is possible using our approach of fusing labels from different annotators, when compared with existing methods presented in the literature; namely two EM models by [1] and [2], as well as a hierarchical Gaussian process approach [8]. The experiments used to validate and compare the methods with selected baselines, along with the results obtained, are detailed before the concluding remarks are presented
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
