Health

An Open Minded Approach To AI Can Improve Our Mental Wellbeing

Technology is blamed for many things. Amongst them is the rise in mental illnesses such as depression, anxiety, and eating disorders. Social media has been linked to a rise in low self esteem and social anxiety for teenagers and adults, and according to psychologist Jean. M. Twenge, technology has resulted in millennials being on the ‘brink of a mental-health crisis.’ Back in 2010 psychologists at the University of Leeds found evidence of a link between excessive internet use and depression, whilst in 2015 scientists at the University of Gothenburg connected the heavy use of technology with heightened stress, sleeping problems and depressive symptoms in women.

But harnessed in the right way, technology offers tremendous opportunity and hope. Research is always pushing boundaries, and HealthTech is set to be one of the most profitable business areas over the next few years. With depression set to be the second largest cause of disability by 2020, according to the World Health Organization, it’s time to consider just how technology can be a force for good.

Machines can help us to detect symptoms

Machine learning is being used to help detect symptoms of mental illness, with a speed and accuracy that humans cannot match. Algorithms have been developed that can identify patterns that are otherwise hidden and hard to detect. Knowledge gathered from such analysis could provide a path to the distinction and recognition of biomarkers – medical signs – present in many illnesses, and potentially offers a more efficient route to diagnosis.

Researchers at the University of Vermont undertook an experiment in which they looked for visual cues in Instagram photos that suggested the subjects are depressed, and then taught a machine how to do the same. They claim that the algorithm is able to identify depression with 70% accuracy. Psychologists at the University of Texas have used a ‘supercomputer’ to identify patterns in neuroimaging data from brain scans that may help predict depression.

Identifying the best treatment options

It may be possible not just to use AI to identify unique types of illness, but also to establish how best to treat them. The use of technology can help predict a patient’s response to certain treatments, thus minimizing the time ‘wasted’ trying other types of intervention. In 2012 Uher and colleagues tested the extent to which different depressive symptoms could predict clinical outcome following 12 weeks of medicinal treatment. They looked for patterns within the datasets which were then generalized to new data generated in later groups of patients. By doing so they were able to predict the efficacy of the treatment. Similarly, a 2016 study by Chekroud and colleagues looked to predict response to a new SSRI, citalopram. They collected the data, then used a machine learning approach to develop a robust algorithm and predictive model.

Self monitoring for information and empowerment

And it’s not just on a big scale within the clinical field that technology can help with mental health. Apps such as Moodtrack Diary, Moodlytics and Mood Tracker can help individuals monitor their emotions and identify patterns or specific circumstances that trigger their depressive moods, whereas others like WellMind, MoodTools and Silvercloud offer activities and strategies for managing depression. They are particularly beneficial where there are personal, social or economic barriers to accessing healthcare, and can also help people to feel empowered to make changes. In the UK, NHS England are pushing more people to manage and engage with their health online in a move to personalize care.

We’re not always great at self monitoring. iSee is an app and tool for clients and counsellors to use together. It combines a GPS tracker, a light sensor, an accelerometer, to capture physical movements, and a touchscreen, to monitor the frequency and duration of users’ interactions with their phones. It takes all of this information and correlates it to patients’ articulation of their moods and emotions, in order to identify what factors increase their vulnerability or resilience. They found that the software ‘provided behavioural markers that were strongly related to depressive symptom severity.’ The idea is that it can help clients and clinicians to manage the depression when they are not in the treatment room.

Reducing waiting times

At the start of the year, the NHS launched a controversial chatbot to help process initial calls to their 111 helpline, aiming to alleviate pressure on the emergency services. Through a mixture of artificial intelligence and video and text consultations with doctors and specialists, patients with non-critical conditions will be offered medical advice and directed to local and out-of-hours medical services. Online cognitive behavioural therapy is now a mainstay in the treatment program for less severe cases, and offers an opportunity to reach more people more easily.

Technology isn’t the whole answer. But neither is it the evil that it is often claimed to be. Advances in the capabilities and the growing availability of digital technologies offers huge opportunities for those with depression. An entrepreneurial and open minded approach that keeps the patient in mind has the potential to hugely improve access, affordability and outcomes.