February 05, 2022

"Machine learning is a type of artificial intelligence that can learn to perform a task independently given enough data and examples. Humans may not have detected these patterns as quickly without the machine."

An expert in machine learning and a psychologist may seem like an unlikely pairing. Researchers at MIT and MGH, Rosalind Picard and Paola Pedrelli, believe that artificial intelligence can help mental health.

A machine can become quite adept at performing a task independently if provided with a large amount of data and examples of good behaviour. As a result, machines can discover meaningful patterns more quickly than humans alone. With the help of wearable sensors and smartphones, Picard and Pedrelli can collect data on the skin conductance and temperature of participants and their heart rate, socialisation levels, self-assessment of depression, and sleep patterns, among other things. Machine learning algorithms can process this massive amount of data and identify when individuals are struggling and what can help them. They hope that their algorithms will one day provide doctors and patients with helpful information about the progression of their disease and the most effective treatment options for them.

Machine learning and mental health

MIT's Jameel Clinic's Picard and Pedrelli are currently working on a project that takes machine learning a step further than the initial focus on determining if you could use data to identify a participant's current emotion. Suppose machine learning can predict the trajectory of a disorder, identify changes in an individual's behaviour, and provide data that you can use to customise medical care. In that case, they are interested in this research.

Consider, for example, a device that can detect when a person has been sleeping less, staying indoors more, or having a higher-than-normal heart rate in the recent past. If these changes are subtle, the individual and their loved ones may not have noticed them. Individuals' and other users' may use prior knowledge and experiences to help machine-learning algorithms make sense of these data.

People can be encouraged to return to healthy habits they've developed over time, or they can use the technology to connect with a doctor.

Empowering users

Apps can be discouraging when they tell a user that they're on the verge of a deep depression, leading to further negative feelings. Actual users are involved in the design process by Pedrelli and Picard to create a tool that is beneficial rather than harmful. Picard suggests. Data privacy and informed consent are also priorities for the team.

Picard claims that using artificial intelligence and machine-learning algorithms, "we can connect and identify patterns in large datasets that humans can't. Using technology to improve one's understanding of others is a compelling argument, in my opinion."

Dr Nivash Jeevanandam PhD,
Researcher | Senior Technology Journalist