5 compelling reasons to choose retained talent recruitment
October 12th, 2021
It is mental health awareness week and thankfully events like these open discussions and help people to share experiences where once there was tumbleweed. Research and understanding have come a long way and now data scientists are helping us understand early detection, diagnosis and treatment much quicker. This blog looks at the latest research and the roles data scientists play in this important area of human life.
The rise of social media has often been cited as having a direct correlation with the rise in anxiety and other mental illnesses, and now data scientists are turning to the medium to analyse the data to help predict and detect early signs of depression. A study in 2017, using machine learning technology, looked at 43,950 participant photos from Instagram to see whether they could successfully identify markers of depression. They used colour analysis of filters used, metadata components and algorithmic face detection and the resulting models outperformed GPs’ average unassisted diagnostic success rate which suggested data science could provide new avenues for early screening and detection of mental illness. Since then, the University of Birmingham has published their research on machine learning and how it can be used to increase the understanding of complex mental illnesses such as depression and psychosis. Many patients present with co-morbid symptoms, e.g. psychosis but also depressive symptoms (for example). Data Scientists discovered through the study that symptoms of depression were much more prevalent in patients presenting with psychosis than clinicians believed. They now know that depression plays a greater part in the illness than had previously been thought. This is a huge breakthrough for more accurate treatments for individuals. The study lead Mr Lalousis explained that “the study highlights the need for clinicians to understand better the neurobiology of these conditions and the role of co-morbid symptoms, in particular considering carefully the role that depression is playing in the illness”.
Data is clearly driving change in diagnostic methods for mental illness and a real-world example of how data predictions have helped save lives can be found in the US based charity, The Crisis Text Line. This is a texting service for people who need to talk to someone about what they are feeling, experiencing or thinking. Sometimes the numbers of people texting in for help far outnumber the capacity of the volunteers to reach them quickly, however the charity has used machine learning to help them prioritise “at risk” individuals. Data Scientists and machine learning have helped them predict which patients need to be pushed up the response queue quicker. They have learned that a crying emoji is 11 times more predictive than the word “suicide” and the words “Advil” and “Ibuprofen” are 14 times more predictive of a suicide attempt than the word “cut”, “die”, “suicide” or “kill”.
They now know, through the role of the Data Scientists on the team, that Wednesday is the most anxiety-provoking day of the week and crises that result in self-harming often happen in the middle of the night. The charity uses this information to initiate “active rescues” where the emergency services are called. The process involved training an algorithm to look at the beginnings of conversations that ended in active rescue. They were able to build a picture, and now the system automatically assigns assistance in order of priority. When spikes happen such as when the pandemic hit, this kind of information is vital for getting to those who need help more urgently.
In the UK the Galatean Risk and Safety Tool (GRiST), developed by Aston University, is a web-based decision support system that helps clinicians to assess and manage risk associated with mental health problems, including suicide, self-harm, self-neglect, vulnerability and harm to others.
There is no doubt that Data Scientists and the models they create through these ongoing studies will have a huge impact on how mental wellbeing is approached in the future. Up until now it has been our physical well-being where Data Science has made the biggest progress, however now the use of Data Science in mental health is finally catching up.
I specialise in Data Science roles and help people carve out a career in this exciting area. I love the difference that this discipline is making to all areas of life and how it can affect decision making. Please do get in touch if you are looking for a new role or need a data scientist in your life. The skills of a talented Data Scientist on your team.
If you need support, Mind Infoline is a confidential service that may be able to help .
Call 0300 123 3393