Researchers from the University of Taiwan claim to have created a digital algorithm which may help to improve early diagnosis of bipolar disorder.

The study comes at a time when mental health charities such as Mind are warning that it can take anywhere up to ten years to diagnose the condition.

Towards the end of 2016, a team of medical professionals from the National Tsing Hua University in Hsinchu City decided to take an unusual approach to slashing this average diagnostic time.

They turned to social media and computer software in a bid to determine whether a person’s online activity could provide an insight into whether they are struggling with their mental health.

Researchers led by Dr Yen-Hao Huang started out by examining patients who were already known to have bipolar disorder.

The team observed the ‘behavioural traits’ of their test subjects online and later published a series of ‘signs’ that friends and other social media users should look out for if they are concerned that someone they know might be depressed.

Using the data, Dr Huang and his associates later created an algorithm which they believe has a 91 per cent accuracy rate when it comes to diagnosing early symptoms of the illness.

“Our experimental results demonstrate that the proposed models could greatly contribute to the regular assessments of people with the condition,” said Dr Huang.

“Identifying the early phases of bipolar disorder is one of the key components for reducing the full development of the disorder,” he added.

According to the NHS, approximately two in every 100 people will encounter bipolar disorder at least once in their lifetime.