The HappyMums Project: Can a smartphone application predict antenatal depression?
- Riddhi Laijawala
- 13 minutes ago
- 4 min read
As a researcher working at the intersection of digital technologies and women’s health, it is always so empowering to see the latest advancements in FemTech (tech-driven products like apps and wearable devices to address female health, like pregnancy and menopause) such as menstrual blood being discovered as a valuable biomarker, and wearable products for menopause detection. It empowers me, as a South Asian woman in science, to do the work I do.
Today, I’m pleased to share that we, as part of the HappyMums project, are contributing to this field. We have recently published the protocol of a clinical study being led by King’s College London, in collaboration with partners across Europe. In our study, we are investigating the use of a smartphone application to gather data to help us predict the development of antenatal depression (during pregnancy).
Why are we doing this?
Antenatal depression affects close to 30% of pregnant women globally, and common risk factors include previous mental health history, lack of social support, and a history of trauma. If left untreated, antenatal depression can have a significant impact on both the mother and baby. It is therefore vital to develop our understanding of the risk factors so that we can identify mothers who may need support to ensure both their well-being, and the well-being of their babies.
Â
What are we going to do?
We are recruiting 1,000 pregnant people between 13 and 28 weeks’ gestation across 7 international centres (approximately 150 per centre), who are either currently suffering with symptoms of antenatal depression, or meet the criteria for at least one risk factor for symptoms. Our screening questionnaire covers aspects such as fertility issues, history of mental health conditions, current consumption of alcohol, and current life stress. Once recruited, consented, and enrolled in the study, they will be given access to the app, for use up to two months after they give birth.
What sets this app apart from the rest?
Brilliant work is being done by others in the field, focused mainly on postpartum depression. Specifically, they are using screening tools to screen for postpartum depression. We contribute to this literature by starting screening for at-risk populations for depression in pregnancy. We wanted to integrate different types of data, not only for screening, but to build machine learning models that will aid clinicians in decision-making in the future.
Our app uses several key features to create a full profile for each participant that gives us information about their ongoing mental health, cognition emotion recognition, and physical activity. These features include:
Mental health questionnaires: Participants will complete validated mental health measures at different timepoints in their pregnancy. This will help us track their mood and mental health across their entire pregnancy.
Games and tasks: different activities, such as a mood and events diary, games to test thinking and memory, and emotion recognition tasks will help us to better understand how each woman is thinking and feeling across their pregnancy.
Smartphone sensors: with the women’s permission we will record their physical activity, such as their step count to understand how and when they move across their pregnancy.
Â
This data will be used to build machine learning models capable of predicting antenatal mental health trajectories. This means that we will harness the benefits of machine learning (a type of artificial intelligence that assists data analysis) to investigate whether it can be useful in seeing which women actually develop depression in pregnancy, from a range of risk factors.
These models will combine multiple data types like the mental health measures and digital data, and the resulting models will be tested for their capabilities of predicting and identifying antenatal depression, as well as response to treatment.
The overall aim is to develop a data collection device which could in future be paired with a dashboard for the patient’s clinician to view their data for use in clinical and treatment decisions.
In addition to this, our participants are given access to a wellbeing course, curated by perinatal experts from the consortium. It contains chapters specific to pregnancy trimesters, covering important topics such as motherhood and biology, birth plans, and breastfeeding, to name a few. We felt it was important to include this, to not only increase the motivation to use the app but also provide relevant, verified information about pregnancy and mental health.
Where are we going to do this?
The HappyMums mobile application study is being conducted at seven recruitment sites across Europe and coordinated by King’s College London. The other six sites are: University of Milan (Italy), Ospedale San Raffaele (Italy), Charité (Germany), University of Helsinki (Finland), SWPS University (Poland) and Catholic University of Croatia (Croatia). As I wrote in a previous article for Inspire the Mind, we have the technological expertise of Abacus (the app developers), and collaborators from the Artificial Intelligence in Medicine Lab at the University of Barcelona. With their help, we will use a federated learning (FL) platform to integrate clinical and digital data from other sites, for analysis. This will allow us to collaborate and share data without relying on legal delays and data transfer agreements. You can read more about this approach here.
I am truly grateful for the opportunity to work on this project. I was once wary of using digital technology in the mental health space, fearing, like many, that artificial intelligence could take away our jobs. Working on this project, however, has made me appreciate the potential artificial intelligence has to aid clinical decision-making. When wielded as a tool, not to replace, but aid clinical decision making, AI has the potential to reduce the burden on healthcare systems, and reduce wait list times. Above all, the work we do is to make sure people get appropriate and timely intervention. With this app, by screening women early on, using a wide range of parameters (digital and clinical), I am excited for the opportunities we have ahead of us.