19 Apr Mass General Study Evaluates AI Narrative to Detect Childbirth-Related PTSD

MedicalResearch.com Interview with:

Dr. Dekel

Sharon Dekel PhD
Principal Investigator
Director of the Postpartum Traumatic Stress Disorders Research Program
Department of Psychiatry, Massachusetts General Hospital
Department of Psychiatry, Harvard Medical School
Boston, MA, 02114

MedicalResearch.com: What is the background for this study?

Response: Maternal psychopathologies affect a significant number of American women and are the leading complications of childbirth and a significant contributor to maternal death. Maternal (physical) morbidity in the US remain the highest among all countries in the West, suggesting that some women will have a traumatic childbirth experience.

The most common mental illness associated with trauma is posttraumatic stress disorder (PTSD). PTSD stemming from childbirth is estimated to affect 6% of delivering women (https://pubmed.ncbi.nlm.nih.gov/28443054/). In high-risk groups, for example women who have unscheduled Cesareans the rate is estimated at 20% or higher (https://pubmed.ncbi.nlm.nih.gov/31041603/.).

Although we screen for postpartum depression in hospitals in the USA there is no screening for what we define as childbirth-related PTSD (CB-PTSD). The overarching goal of the Dekel Lab is to develop novel and patient-friendly screening tools to identify women with this disorder. As importantly traumatic childbirth disproportionality affects Black and Latina women (https://pubmed.ncbi.nlm.nih.gov/35598158/).

MedicalResearch.com: What are the main findings?

Response: In this study, we examined whether brief narrative accounts of childbirth in which women write about the most distressing aspects of their experience can be an avenue into their psychological well-being. For this purpose, we used large language AI models and text embedding that represented the language in vectors. These embeddings then served as an input of a machine learning model that was trained on a set of narratives of women with and without CB-PTSD and then tested to identify if the person has CB-PTSD based on a new narrative. Our findings show that the model has strong capabilities in classifying women correctly.

MedicalResearch.com: Is this AI model specific for post childbirth PTSD?  Can it be amended/updated for other mental health issues?

Response:  The AI model used birth narrative, potentially traumatic accounts of childbirth to detect PTSD. It is possible that this model could be used to identify other forms of PTSD stemming from other types of traumas when used with narratives of survivors. In general, big data such as information (free text) in social media analyzed by large language models may offer promising methods to inform the efficient screening of mental illnesses.

MedicalResearch.com: Is there anything else you would like to add? Any disclosures?

Response: Dr. Dekel’s work is funded by the NIH, NICHD (R01HD108619,  R21HD109546, R21HD100817). The work was performed in collaboration with Dr. Alon Bartal from Bar Ilan University in Israel. Pending more funding, the study team will continue and develop their model in culturally diverse postpartum populations. Screening is key for effective treatment and may serve as the first step in preventing CB-PTSD (rhttps://pubmed.ncbi.nlm.nih.gov/38122842/).

To learn about the Dekel Lab and contribute to our mission to optimize maternal mental health visit: https://massgeneral.link/DekelLab

Citation: Bartal, A., Jagodnik, K.M., Chan, S.J. et al. AI and narrative embeddings detect PTSD following childbirth via birth stories. Sci Rep 14, 8336 (2024). https://doi.org/10.1038/s41598-024-54242-2

The information on MedicalResearch.com is provided for educational purposes only, and is in no way intended to diagnose, cure, or treat any medical or other condition.

Some links may be sponsored. Products are not endorsed.

Always seek the advice of your physician or other qualified health and ask your doctor any questions you may have regarding a medical condition. In addition to all other limitations and disclaimers in this agreement, service provider and its third party providers disclaim any liability or loss in connection with the content provided on this website.

Last Updated on April 19, 2024 by Marie Benz MD FAAD