Synthetic Subject Generation with Coupled Coherent Time Series Data

Larrea, Xabat; Hernandez, Mikel; Epelde, Gorka; Beristáin, Antonio; Molina, Cristina; Alberdi, Ane; Rankin, Debbie; Bamidis, Panagiotis D.; Konstantinidis, Evdokimos I.

doi:10.3390/engproc2022018007

Cited by 2 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main contribution of this work is the detailed definition and systematic comparison of the three methods for generating time series together with metadata in terms of different evaluation dimensions, presenting them in a common framework and substantially extending analyses previously published by the authors in [ 12 , 13 ].…”

Section: Discussionmentioning

confidence: 99%

“…First, the TMET dataset was preprocessed following the steps defined in Larrea et al [ 12 ]. Starting from a 992-subject dataset, 30 of them were directly excluded due to missing values on subject metadata.…”

Section: Methodsmentioning

confidence: 99%

“…In this study, three different approaches to generate synthetic time series are presented and compared, namely synthetic metadata generation and real time series coupling or A1, separate synthetic metadata and time series generation and later coupling or A2, and joint synthetic metadata and time series generation, A3. The first two by Larrea et al [ 12 ] were earlier introduced in the manuscript, as well as the third approach considered by Isasa et al in [ 13 ]. In this section, more detailed explanations about the different approaches can be found supported by Fig.…”

Section: Methodsmentioning

confidence: 99%

“…In 2022, we proposed two different approaches for the Synthetic Time Series Generation (STSG) task [ 12 ], where different combinations of real and synthetic metadata and time series are merged to find the best approach for achieving useful time series. In the present work, a more extensive evaluation of both approaches is carried out assessing the resemblance of the SD with respect to the RD (1), how useful the SD is to build downstream applications (2), and how private the generated information is (3).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Comparative assessment of synthetic time series generation approaches in healthcare: leveraging patient metadata for accurate data synthesis

Isasa,

Hernandez,

Epelde

et al. 2024

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

Background Synthetic data is an emerging approach for addressing legal and regulatory concerns in biomedical research that deals with personal and clinical data, whether as a single tool or through its combination with other privacy enhancing technologies. Generating uncompromised synthetic data could significantly benefit external researchers performing secondary analyses by providing unlimited access to information while fulfilling pertinent regulations. However, the original data to be synthesized (e.g., data acquired in Living Labs) may consist of subjects’ metadata (static) and a longitudinal component (set of time-dependent measurements), making it challenging to produce coherent synthetic counterparts. Methods Three synthetic time series generation approaches were defined and compared in this work: only generating the metadata and coupling it with the real time series from the original data (A1), generating both metadata and time series separately to join them afterwards (A2), and jointly generating both metadata and time series (A3). The comparative assessment of the three approaches was carried out using two different synthetic data generation models: the Wasserstein GAN with Gradient Penalty (WGAN-GP) and the DöppelGANger (DGAN). The experiments were performed with three different healthcare-related longitudinal datasets: Treadmill Maximal Effort Test (TMET) measurements from the University of Malaga (1), a hypotension subset derived from the MIMIC-III v1.4 database (2), and a lifelogging dataset named PMData (3). Results Three pivotal dimensions were assessed on the generated synthetic data: resemblance to the original data (1), utility (2), and privacy level (3). The optimal approach fluctuates based on the assessed dimension and metric. Conclusion The initial characteristics of the datasets to be synthesized play a crucial role in determining the best approach. Coupling synthetic metadata with real time series (A1), as well as jointly generating synthetic time series and metadata (A3), are both competitive methods, while separately generating time series and metadata (A2) appears to perform more poorly overall.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations