TY - JOUR
T1 - Synthetic data & the future of Women's Health
T2 - A synergistic relationship
AU - Delanerolle, Gayathri
AU - Phiri, Peter
AU - Cavalini, Heitor
AU - Benfield, David
AU - Shetty, Ashish
AU - Bouchareb, Yassine
AU - Shi, Jian Qing
AU - Zemkoho, Alain
N1 - Funding Information:
The authors acknowledge administrative support from Sana Sajid and Southern Health NHS Foundation Trust, Southern University of Science and Technology and University of Southampton.
Publisher Copyright:
© 2023 Elsevier B.V.
DBLP License: DBLP's bibliographic metadata records provided through http://dblp.org/ are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.
PY - 2023/11/1
Y1 - 2023/11/1
N2 - Objectives: The aim of this perspective is to report the use of synthetic data as a viable method in women's health given the current challenges linked to obtaining life-course data within a short period of time and accessing electronic healthcare data. Methods: We used a 3-point perspective method to report an overview of data science, common applications, and ethical implications. Results: There are several ethical challenges linked to using real-world data, consequently, generating synthetic data provides an alternative method to conduct comprehensive research when used effectively. The use of clinical characteristics to develop synthetic data is a useful method to consider. Aligning this data as closely as possible to the clinical phenotype would enable researchers to provide data that is very similar to that of the real-world. Discussion: Population diversity and disease characterisation is important to optimally use data science. There are several artificial intelligence techniques that can be used to develop synthetic data. Conclusion: Synthetic data demonstrates promise and versatility when used efficiently aligned to clinical problems. Therefore, exploring this option as a viable method in women's health, in particular for epidemiology may be useful.
AB - Objectives: The aim of this perspective is to report the use of synthetic data as a viable method in women's health given the current challenges linked to obtaining life-course data within a short period of time and accessing electronic healthcare data. Methods: We used a 3-point perspective method to report an overview of data science, common applications, and ethical implications. Results: There are several ethical challenges linked to using real-world data, consequently, generating synthetic data provides an alternative method to conduct comprehensive research when used effectively. The use of clinical characteristics to develop synthetic data is a useful method to consider. Aligning this data as closely as possible to the clinical phenotype would enable researchers to provide data that is very similar to that of the real-world. Discussion: Population diversity and disease characterisation is important to optimally use data science. There are several artificial intelligence techniques that can be used to develop synthetic data. Conclusion: Synthetic data demonstrates promise and versatility when used efficiently aligned to clinical problems. Therefore, exploring this option as a viable method in women's health, in particular for epidemiology may be useful.
KW - Electronic health records
KW - Machine learning
KW - Real-world Data
KW - Synthetic data
KW - Women's Health
KW - Health Services Accessibility
KW - Humans
KW - Artificial Intelligence
KW - Female
UR - http://www.scopus.com/inward/record.url?scp=85173286317&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85173286317&partnerID=8YFLogxK
UR - https://www.mendeley.com/catalogue/bfccfbea-9730-3146-bc42-55054f723887/
U2 - 10.1016/j.ijmedinf.2023.105238
DO - 10.1016/j.ijmedinf.2023.105238
M3 - Article
C2 - 37813078
AN - SCOPUS:85173286317
SN - 1386-5056
VL - 179
SP - 105238
JO - International Journal of Medical Informatics
JF - International Journal of Medical Informatics
M1 - 105238
ER -