Bayesian fusion of physiological measurements using a signal quality extension

Abstract

Objective: The fusion of multiple noisy labels for biomedical data (such as ECG annotations, which may be obtained from human experts or from automated systems) into a single robust annotation has many applications in physiologic monitoring. Directly modelling the difficulty of the task has the potential to improve the fusion of such labels. This paper proposes a means for the incorporation of task difficulty, as quantified by ‘signal quality’, into the fusion process. Approach: We propose a Bayesian fusion model to infer a consensus through aggregating labels, where the labels are provided by multiple imperfect automated algorithms (or ‘annotators’). Our model incorporates the signal quality of the underlying recording when fusing labels. We compare our proposed model with previously published approaches. Two publicly available datasets were used to demonstrate the feasibility of our proposed model: one focused on QT interval estimation in the ECG and the other focused on respiratory rate (RR) estimation from the photoplethysmogram (PPG). We inferred the hyperparameters of our model using maximum- a posteriori inference and Gibbs sampling. Main results: For the QT dataset, our model significantly outperformed the previously published models (root-mean-square error of ms for our model versus ms from the best existing model) when fusing labels from only three annotators. For the RR dataset, no improvement was observed compared to the same model without signal quality modelling, where our model outperformed existing models (mean-absolute error of bpm for our model versus bpm from the best existing model). We conclude that our approach demonstrates the feasibility of using a signal quality metric as a confidence measure to improve label fusion. Significance: Our Bayesian learning model provides an extension over existing work to incorporate signal quality as a confidence measure to improve the reliability of fusing labels from biomedical datasets.

Atrial fibrillation detection in outpatient electrocardiogram monitoring: An algorithmic crowdsourcing approach

“…Various fusion mechanisms or voting strategies are proposed in the literature ranging from naïve voting system based on the best performance of algorithms, to LASSO, LASSO+ [ 16 ], and Bayesian approaches [ 18 – 21 ], to combine the algorithms’ outputs in smarter and systematic ways. In this work, we examined multiple bagged and boosted decision trees to combine different algorithms, including bagged decision tree, random forest, AdaBoost [ 22 ], RUSBoost [ 23 ], and TotalBoost [ 24 ].…”

Machine learning from crowds: A systematic review of its applications

“…Other interesting applications in this area include: estimation of respiratory rate from the gene normalization (Lu et al, ); estimation of fetal heart rate, interbeat intervals and fetal QT intervals with noninvasive electrocardiograms (ECG) (Silva et al, ); protein folding (Peng et al, ); ECG signal classification (Zhu et al, ; Zhu, Dunkley, et al, ); photoplethysmograms (Zhu, Pimentel, et al, ); sleep spindle detection (Tan et al, ); assessment of voice pathologies (González et al, ) and learning for ICD‐11 sanctioning rules (Lou et al, ); labeling respiratory patterns (Robles‐Rubio et al, ); analysis of respiratory data (Zhu et al, ); and learning rules for disease‐remedy associations (Someswar & Bhattacharya, ).…”

Classifying Tremor Dominant and Postural Instability and Gait Difficulty Subtypes of Parkinson’s Disease from Full-Body Kinematics

“…To avoid subjective MDS-UPDRS-III scoring as a gold standard measure of movement, at least two trained movement disorder specialists should examine the patient with inter-rater reliability over 90% [13,14]. When having multiple raters, assessment results can be further improved with techniques like Bayesian aggregation [15]. However, employing multiple raters is very costly with limited resources and accessibility to medical services globally [16,17].…”