How AI Can Help Solve the Growing Mental Health Crisis

How AI Can Help Solve the Growing Mental Health Crisis

Aiding mental health with a touch of technology.

After years of research, artificial intelligence (AI) has become a part of our society. From automated customer service representatives to cars that can drive themselves on the road. And now, the technology is being used to address one of the world’s most pressing concerns: mental health. 

A survey by the World Health Organization shows that the Covid-19 pandemic has disrupted or halted critical mental health services in 93 percent of countries worldwide, and the demand for mental health care is increasing. Bereavement, isolation, loss of income and fear are all triggering mental health problems. Due to the lockdowns imposed in the country, people in distress could not access mental healthcare professionals. Here, technology has stepped in, with people using mental healthcare apps to seek comfort in such troubled times.

Here we list down a few ways AI and other technologies are revolutionizing the mental health industry.

Preventing suicides and self-harm

Machine learning (ML), a kind of AI, has the potential to better detect suicidal risks in human beings. With a more efficient suicidal ideation identification process, the transmission of relevant information to the people at risk can also be much quicker.

Tech giants, like Facebook, have expanded their abilities to aid people in need with ML. These ML tools analyze the words and phrases in posts and comments from friends and family to identify potential suicidal threats. When the system identifies such content in a post, members from the Community Operations team will determine if the user is at risk of self-harm and whether there is a need to take further actions.

Another example of adopting ML to improve mental health services is the Crisis Text Line, a 24/7 text messaging-based crisis hotline that collects data of people experiencing despair. During the pandemic, calls to mental health hotlines have been increasing. This could put immense pressure on frontline mental health workers, making it difficult for them to help people in despair in time. The Crisis Text Line uses ML to pull out words or emojis that indicate suicide ideation. The data collected by the helpline has revealed various insights on how to help people through tough times. For instance, it finds out that Wednesday is the most anxiety-provoking day of the week. 

Predicting depression 

Researchers are assessing various ways AI can help screen, diagnose and treat mental health problems. Researchers from the World Well-Being Project (WWBP) examined social media data with an AI algorithm to pull out linguistic signs that might predict depression. The researchers identified a range of language markers that could predict depression up to three months before a person is formally diagnosed. 

Besides text, voice can also be helpful in detecting depression. Sonde Health is training ML models that can give cues when people begin to experience depressive symptoms by listening to them talk. The CEO of Sonde Health, David Liu, says, “When we listen to a person speaking, we notice variations in pitch, energy, tonal quality, and rhythm……By processing this audio, we can break down a few seconds of voice recording into a signal with thousands of unique characteristics.” This method, which is called audio signal processing, can identify changes in vocal features that are common among people with depression.

Assisting mental health treatment

Besides identification, AI technologies, such as natural language processing (NLP), could be used in therapist-driven mental health treatment. The Natural Language Processing Group at Stanford University utilizes data mining and NLP to analyze over 80,000 counseling sessions and extract information from text. The project’s objective was to determine how to conduct better counseling sessions. To do this, the researchers identified linguistic elements in conversations that indicated the patients had felt better afterward. It found that the more successful counselors tend to adopt these five strategies: adaptability (reacting to the conversation accordingly), dealing with ambiguity (clarifying and affirming the situation with patients), creativity (responding creatively), making progress (moving on to solving problems quickly) and change in perspective (bringing up positive concepts). These findings can help enhance counselor training and lead to tools to improve counseling sessions.

AI is already being used to improve mental health through apps and programs. Although technology does come with certain privacy concerns, AI might be the solution to the pandemic-borne mental health crisis with the continuous developments in healthcare.

Header Image by Unsplash


Share on facebook
Share on twitter
Share on linkedin
Share on email