Responsible Use of AI in Obstetric Care: Navigating Ethical Considerations as a Healthcare Leader

By John Parker, MD, FACOG, Chief Medical Officer

Artificial intelligence (AI) has emerged as one of the most transformative technologies of the 21st century. From improving diagnostic accuracy to personalizing treatment plans and streamlining operations, its potential to revolutionize obstetric care is immense. However, as this technology evolves, so too must our knowledge of how to use it ethically and responsibly.

For healthcare leaders, understanding how AI works is essential. As you become more involved in evaluating AI-powered products and services for patient care, the ability to critically assess the type of AI being used, how it was trained and validated, and its potential for bias is becoming more important than ever. By prioritizing and advocating for privacy, transparency, and fairness, you can play a vital role in ensuring your health system leverages AI responsibly and effectively.

This blog explores the ethical principles that should guide the secure, transparent, and equitable use of AI in obstetric care, as well as how PeriGen is applying these principles.

Data Privacy and Security: The Foundation of Responsible AI

What You Need to Know: AI systems rely on large volumes of sensitive health data. Without strong safeguards, that data can be misused, exposed, or accessed by unauthorized individuals — putting patient privacy, institutional reputation, and regulatory compliance at risk. As AI becomes more integrated into care, healthcare leaders must ask critical questions about data privacy — how patient information is protected, who can access it, and whether the system complies with regulatory and organizational security standards.

How PeriGen Supports Data Privacy and Security: PeriGen collaborates with hospital IT departments to implement single sign-on (SSO) capabilities, enhancing the end-user experience while aligning with each health system’s access control policies. This allows health systems to ensure that only authorized personnel can access the data used or generated by the software. In addition, PeriGen’s products are FDA-cleared Class II medical devices and adhere to the FDA’s rigorous cybersecurity requirements, further supporting safe and secure use in clinical environments.

The Importance of Transparency in Healthcare AI

What You Need to Know: Understanding how AI works is essential for building trust in healthcare. While many people associate AI with generative tools like ChatGPT, these continuously evolving models can pose risks when applied to clinical care. In contrast, supervised machine learning — a subset of AI that uses labeled datasets to train algorithms, followed by expert validation — can offer a more reliable approach. Because this form of AI is static and does not continue to evolve after its initial training, it may be more appropriate for certain healthcare applications.

For obstetric leaders, it’s important to have transparency into how any AI tool you want to incorporate into patient care was developed. Ask vendors how their algorithms were trained and validated, and request an AI model card if available. A model card outlines the intended use, training and evaluation data, quantitative analyses, ethical considerations, and any known limitations — helping you make informed, responsible decisions.

How PeriGen Applies Transparency: PeriGen uses a supervised machine learning algorithm to interpret fetal heart rate patterns and detect uterine contractions. The model was trained in the lab on over 1,800 hours of carefully curated tracings representing a wide range of clinical conditions. Its performance was validated in a study submitted to the FDA as part of the clearance process. Once trained, the algorithm is static and does not continue learning in the clinical environment.

Bias in, Bias Out: Why Diverse Training Data and Validation Matter

What You Need to Know: AI systems are only as reliable as the data used to train them. When training data lacks diversity, it can lead to biased outcomes that disproportionately impact certain patient groups. For instance, if an AI model is trained primarily on data from one demographic, its performance may suffer when applied to others. To ensure fairness and equity in healthcare, it’s essential to use diverse training datasets and implement robust strategies for bias detection and mitigation.

How PeriGen Mitigates Bias: PeriGen’s algorithm does not use demographic data in its decision-making. Fetal heart rate (FHR) interpretation is standardized for all maternal/fetal profiles beyond 32 weeks of gestation. To confirm the model’s fairness, the training data was stratified by key clinical factors — such as gestational age, maternal BMI, Apgar score, umbilical cord pH, birthweight, and acquisition device — to ensure these variables did not affect the output. As an FDA-cleared medical device, PeriGen continues to monitor for potential bias through ongoing post-market surveillance activities.

Your Role in Shaping the Future of Care

As artificial intelligence continues to shape the future of obstetric care, the role of healthcare leaders in guiding its responsible adoption becomes increasingly vital. From safeguarding patient privacy to ensuring algorithm security, transparency, and fairness, your voice is critical in evaluating and championing technologies that align with clinical values and ethical standards.

At PeriGen, we are committed to supporting and collaborating with healthcare leaders who are guiding the thoughtful integration of AI into obstetric care. By working together, we can harness the power of AI to elevate the standard of care for mothers and babies everywhere.

Originally published on synovaassociates.com.