Mom2B: Predicting perinatal depression using smartphone-based digital phenotyping and maching learning

Main Supervisor: Fotis Papdopoulos
Co-upervisor: Ginevra Castellano, Stavros Iliadis, and Caisa Öster

What is your educational background?
I have a BA in psychology, and a MSc in Clinical Psychology, with research experience in health psychology. Furthermore, I have worked as a research assistant in a project promoting early intervention for mother-infant bonding among refugee women experiencing postnatal depression.

Why did you apply to WOMHER's interdisciplinary graduate school?
Interdisciplinary research is at the forefront of academia now and is a great opportunity to expand the boundaries of research in any one specific field. My main motivation for applying to WOMHER was the nature of the research project. I had been working in psychology whilst having a strong personal interest in technology and artificial intelligence. It was important for me that the project I applied was keeping up with the times and incorporating new, advanced methods we have available today to collect data and analyze it. As part of WOMHER, I was also ablet to meet other researchers working in the area of women’s mental health, which is not only a great networking opportunity, but also a helpful support network to have.

Tell us more about your research project?
Mom2B is a smartphone app-based study. The aim of this project is to develop and evaluate machine learning and deep learning models predicting the risk of depression in the perinatal period. Women are able to download the Mom2B app and provide data via the app. The project looks at episodes of depression anytime during pregnancy and up to one year postpartum. One of the novelties of this project is our collection of phone sensor data to infer behavioral patterns in everyday life, in addition to self-reported information. Furthermore, we aim to explore the acceptability and experience of the app for users to improve our understanding of how mobile health apps like this can be best implemented in the real world.

What do you hope the impact of this project to be?
My PhD places a lot of focus on the implantation and acceptability of this app and the algorithm in real world settings. For me, it would be great to see the algorithm we develop be used in the form of a self-help app, or an app that can be integrated into the healthcare system and help detect women at risk of depression so that they can be offered early preventative interventions. Being able to track one’s own health privately and personally should be empowering for women, and break some of patient-level barriers to screening and seeking help.

Other information, references and links
Predicting perinatal health outcomes using smartphone-based digital phenotyping and machine learning in a prospective Swedish cohort (Mom2B): study protocol. BMJ Open, 12(4). DOI: http://dx.doi.org/10.1136/bmjopen-2021-059033