Exploring machine learning and AI for self-care in menopause

Pausetrack is an app that helps women in peri-menopause to understand their symptoms and self-care. Pausetrack can track periods and up to 25 symptoms of peri-menopause, and can also track 'interventions' or the things that women try to help themselves. We are collecting big data that we hope to use to inform better research, policy and practice. In future versions of the app when we have big data we will use AI and machine learning to understand peri-menopause and inform better care solutions.

Our users are recording more symptoms then interventions and our user feedback suggests this is because women are not sure what to try to help themselves. This project aims to exploit this gap and offer a bridge between big data, clinical expertise and users by automating analysis of symptoms and interventions through the app. Users when recording a symptom can then be presented with a simple pop up displaying personalised evidence-based intervention recommendations that might work for them. For example if a user records insomnia in the app a pop-up box will appear showing 'Users like you find these \[interventions\] improved their sleep'. The user will then be offered links to try the intervention and record it in the app.

Behind the scenes machine learning will look for patterns in the data we are collecting and build a database so that we can share this information with users. We will ask different health professionals (GP, Pharmacist, Nurse) to look at our data and models to check clinically they are good quality and relevant. This information will then be transferred to the database and interpreted into a format that can be used by the app to create the pop-up for users.

Using clinically validated big data to make personalised self-care suggestions is a novel approach but which could have big benefits for users. It is often difficult for the public to have access to and understand good quality personalised health suggestions that could help them in their day to day lives. We hope to resolve this issue for women in peri-menopause and give them direct access to personalised solutions derived through automated research analysis.