PlethAugment: GAN-Based PPG Augmentation for Medical Diagnosis in Low-Resource Settings

Abstract

The paucity of physiological time-series data collected from low-resource clinical settings limits the capabilities of modern machine learning algorithms in achieving high performance. Such performance is further hindered by class imbalance; datasets where a diagnosis is much more common than others. To overcome these two issues at low-cost while preserving privacy, data augmentation methods can be employed. In the time domain, the traditional method of time-warping could alter the underlying data distribution with detrimental consequences. This is prominent when dealing with physiological conditions that influence the frequency components of data. In this paper, we propose PlethAugment; three different conditional generative adversarial networks (CGANs) with an adapted diversity term for the generation of pathological photoplethysmogram (PPG) signals in order to boost medical classification performance. To evaluate and compare the GANs, we introduce a novel metric-agnostic method; the synthetic generalization curve. We validate this approach on two proprietary and two public datasets representing a diverse set of medical conditions. Compared to training on non-augmented class-balanced datasets, training on augmented datasets leads to an improvement of the AUROC by up to 29% when using cross validation. This illustrates the potential of the proposed CGANs to significantly improve classification performance.

Generating Synthetic Mixed-type Longitudinal Electronic Health Records for Artificial Intelligent Applications

“…EHRs are sets of digital patient-centered records collected over time, thereby recording patient health status and care trajectories during the hospitalization. Recent research has aimed to use this information to generate clinical timeseries which capture the character of these trajectories [30][31][32][33]. In this study, we focus on generating timeseries in critical care, specifically in intensive care units (ICUs) where patient information are fully digitized.…”

Generative adversarial networks in time series: A survey and taxonomy

“…It is common knowledge in the deep learning community that GANs are among the methods of choice when discussing data augmentation. Reasons for augmenting datasets range from increasing the size/variety of small and imbalanced datasets [62,63,64,65] to reproducing restricted datasets for dissemination.…”

Generating Synthetic Mixed-type Longitudinal Electronic Health Records for Artificial Intelligent Applications

“…We partition the dataset as illustrated in Figure 6a, and the performances are assessed from two aspects (see Figure 6b): (i) Traditional approach: To explore whether the synthetic data can represent the real data accurately, we compare Train on Real, Test on Real (TRTR) with Train on Synthetic, Test on Real (TSTR), to show whether the performance of a classifier trained on synthetic data from EHR-M-GAN or EHR-M-GAN cond can be generalized to real data; and (ii) Data augmentation approach: As data augmentation is employed as a means of circumventing the issue caused by the under-resourced EHR data, here we explore whether synthetic data can used to improve the existing ML algorithms through data augmentation. Therefore, Train on Synthetic and Real, Test on Real (TSRTR) is compared with TRTR to measure the improvement of the classifier's performance when trained on the augmented data (Esteban et al, 2017;Kiyasseh et al, 2020). The augmentation ratio α or β is applied on sub-train data A T r or synthetic data B, in two different scenarios of TSRTR, respectively.…”

“…EHRs are sets of digital patient-centered records collected over time, thereby recording patient health status and care trajectories during the hospitalization. Recent research has aimed to use this information to generate clinical timeseries which capture the character of these trajectories (Esteban et al, 2017;Lee et al, 2020;Zhang et al, 2021;Kiyasseh et al, 2020). In this study, we focus on generating timeseries in critical care, specifically in intensive care units (ICUs) where patient information are fully digitized.…”