Skip to main content
SMFM 2024 Pregnancy Meeting Main banner
Final Program


Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
Additional registration fee required
Additional registration fee required
Faculty have requested this content not be shared on social media
Faculty have requested this content not be shared on social media
CME Credit Offered
CME Credit Offered
Poster Icons
Foundation Awardee
Foundation Awardee
No Social Media
No Social Media
Back
Q&A

Hypertension

Poster Session 4

(1192) The Evolving Landscape of Biomedical Informatics for Postpartum Hypertension: A Scoping Review

Wednesday, February 14, 2024
1:00 PM – 2:30 PM  
Location: Prince George's Exhibit Hall CDE, Lower Atrium Level
Foundation Awardee
Objective: Postpartum hypertension (PPHTN) affects more than 10% of deliveries and can affect multiple systems such as neurological and renal systems in the mother while also affecting the infant negatively. The objective of this scoping review is to investigate the role of biomedical informatics in the identification and management of PPHTN, with a focus on the use of machine learning/artificial intelligence (ML/AI) for early identification.

Study Design: A structured search was used to query Scopus, PubMed, IEEE Xplore for relevant articles. The scoping review identified studies that described the application of ML/AI for prediction, prevention, or early detection of adverse events in each of the harm domains. Inclusion criteria: Included PPHTN or high blood pressure occurring within the first six weeks following childbirth, studies based on humans, ML studies with PPHTN, and ML studies in postpartum related complications. Exclusion criteria removed papers unrelated to the postnatal period, meta-analyses, any review, prenatal or gestational only studies.

Results:

A total of 462 articles were obtained with the keywords “machine learning”, “artificial intelligence”, with the following keywords “postpartum hypertension”, “high blood pressure”, “postnatal hypertension”. Four articles were selected after screening according to the inclusion and exclusion criteria. Data used to predict PPHTN were electronic health records (EHR) (75%), medical records (25%). In 3 studies, the main metrics used were: Area under the curve, sensitivity, specificity, and accuracy. Most common ML methods used were logistic regression and XGBoost. Out of the ML approaches used, the best results were found using XGBoost technique with an AUC score of 0.85.

Conclusion:

Our results demonstrate that AI-based methods to reduce PPHTN have been limited to a small number of studies that focus on structured EHR data elements. In conclusion, our findings highlight the potential for AI to improve diagnosis and treatment of patients with PPHTN by overcoming various problems related to early identification, and management of postnatal hypertension.