Clinical Research

AI-Driven Maternal Biometric Analysis for Pregnant Women

Background

Preterm birth and stillbirth are major social and public health challenges worldwide. While a range of factors—such as maternal age, infectious diseases, and genetic abnormalities—have been identified as potential causes, many aspects remain unclear.
Amid broader social changes, particularly the rise in pregnancies at advanced maternal age, the risk of preterm birth is increasing.
In 2020, the World Health Organization (WHO) sounded the alarm on the state of preterm birth, describing its global burden as a “silent emergency,” noting that rates have not improved in decades.
The impact of preterm birth and stillbirth extends beyond individuals to society at large. The costs associated with tertiary care and subsequent lifelong support are substantial, with the United States spending more than USD 20 billion annually on the management of preterm birth and the United Kingdom spending approximately GBP 2.5 billion.

Research Overview

Guided by medical and physiological perspectives, we hypothesize that there is a mapping between multimodal maternal biometric data and the intrauterine environment—and that these relationships can be modeled and inferred using deep learning.
Since our founding, we have conducted multiple research projects and completed several proofs of concept (PoCs). Building on the knowledge gained along the way, we are currently developing a foundation model for pregnancy.
As a future application of this model, we are in discussions with a national university to conduct clinical research that explores the potential for early intervention in high-risk cases such as preterm birth and stillbirth.

Research Outcomes

At the 2025 Annual Congress of the Japan Society of Obstetrics and Gynecology (JSOG), we presented our findings to medical experts. We are currently preparing a manuscript for submission to a peer-reviewed journal.
We will update this page as new results become available.

Contact

If you are a study participant and have any questions, please contact us via the form specified in the informed consent document.
To protect participants’ personal information, we are unable to respond to individual inquiries about this clinical research submitted through the website’s general “Contact Us” form. We appreciate your understanding. For details on how we handle personal information, please refer to our privacy policy at the link below.

References