Could AI Understand Mental Health Better Than We Do?

Computer brains may help us to understand our own

It is often said that the human brain is the most complex structure in the universe. Our brains contain over 100 billion neurons which sit in a complicated web of 100 trillion possible connections. There are more nerves in the brain than there are stars in our galaxy – if you laid them out end to end they would stretch round the Earth twice. In an organ of such vast complexity, it is no wonder that things sometimes go wrong. The treatment of brain disorders and mental health problems relies upon our understanding and accurate diagnosis of these issues. AI technology is now helping us to do just that.

Mental health problems are a crisis of our modern age. In the US alone, one in five adults currently experience a mental health disorder such as depression, schizophrenia, and bipolar disease. For employers it’s important to take care of your employees both from a humanitarian perspective and also a business one. In fact, poor mental health costs UK employers between £33 billion and £42 billion a year, and the global economic cost of mental health problems is expected to reach USD $6 trillion by 2030. There are lots of things that employers can do to help staff going through a rough time. But to really improve the state of global mental health we need to better understand the brain. The computing power of artificially intelligent technology may provide the answer.

Our speech patterns can reveal disorders and mental health problems

IBM has developed an automated speech analysis programme to find early indicators of mental illness. By combining text to speech software, advanced analytics, machine learning, natural language processing technologies and computational biology, they have created an application for mobile devices which can analyse the way we talk. This impressive technology is able to pick out patterns in a patient’s speech or written words to assess their meaning, syntax and intonation – all of which can provide insights into a person’s mental health. Once collected, data from the programme can be combined with that from wearables and imaging devices such as MRIs, to build up a picture of the individual. Artificially intelligent technology then analyses this data to aid medical professionals in their diagnosis, treatment and monitoring. This technique is able to pick up conditions including depression, degenerative neurological diseases such as Parkinson’s and developmental disorders like ADHD.

This process of assessment and analysis is nothing new in healthcare, including psychiatry. However, without the appropriate technology it is extremely labour intensive to survey a patient in this way. Multiple sessions with an individual are required to conduct and record interviews manually, and then comes the significant challenge of analysing the data. Artificial Intelligence, on the other hand, can easily find patterns and draw conclusions that a human worker would miss. Their strength of data analysis is a powerful tool, especially given the fact that many mental health problems currently remain unnoticed and undiagnosed.

100 percent accuracy of predictions

In a 2015 study with Columbia University, IBM’s software was able to predict with 100 percent accuracy which members of a group of at risk adolescents would develop their first episode of psychosis within two years. Providing objective clinical tests such as these is a real landmark in psychiatry, as it enables clinicians to assess their patients in ways which would previously have been impossible. As a matter of fact, IBM states that it only takes around 300 words of analysed speech for clinicians to predict the probability of psychosis in a patient. Such concrete, measurable diagnostic terms are of great value to a specialisation which has previously lacked them. It enables early identification of disease, giving patients the best chance of recovery.

Speech analysis technology such as this is good news for all of us, from those suffering from the most severe mental disorders, to the average person looking to keep an eye on their mental health. Any way that we can foster good mental health practice will not only be beneficial to us as individuals, but will help us to unlock greater economic potential on a global scale.

In what way is your business affected by poor mental health? Can we afford to ignore this growing crisis? Would you use intelligent technology to keep track of your personal wellbeing? Comment below with any thoughts or opinions.

For more insights from D/SRUPTION you can sign up for our FREE weekly newsletter here.