Big Data, Big Psychiatry = Big Potential
, Big Data, Big Psychiatry — Big Potential?
• Big Data is nearly impossible to manage with traditional software and hardware tools.
• Computers can learn to spot patterns (meaning) in large sets of data much more quickly and
reliably than humans.
• Requires solutions based on pattern recognition (process of machine learning) whereby
computers identify complex interactions among different sets of variables.
• Improving healthcare – Data-driven medicine involves analysing vast numbers of medical
records and images for patterns that can help spot disease early and develop new
medicines.
• Big Data could immensely aid in the management and analysis of observational data.
• Combining both social media postings and clinical data can uncover behavioural patterns
that are otherwise hard to detect by other means.
Using Social Media (Twitter) To Predict Depression
• Social media provides an important space for many people dealing with rare and often
stigmatized mental illnesses.
• Many individuals turn to Facebook groups or online gaming platforms to connect with others
who share similar experiences.
• What do twitter profiles and posted images reveal about depression and anxiety?
• Ungar et al., 2019 found that the colour, lighting and symmetry of photos shared on Twitter
may give insight into who is at risk for anxiety and depression.
• Those that had depression were less likely to have a profile picture that was happy or that
exhibited a particular mood
• Demonstrates the use of pictures as data.
• https://arxiv.org/pdf/1904.02670.pdf
Using big data to test whether a man and women will (unite) go on a second date
Can you predict whether a man and women will go on a second date based on how they speak on
the first date?
• The authors used traditional data (height, weight, hobbies), together with recordings of the
conversations.
• The scientists coded the words used, the presence of laughter and tone of voice.
, Big Data, Big Psychiatry — Big Potential?
• Big Data is nearly impossible to manage with traditional software and hardware tools.
• Computers can learn to spot patterns (meaning) in large sets of data much more quickly and
reliably than humans.
• Requires solutions based on pattern recognition (process of machine learning) whereby
computers identify complex interactions among different sets of variables.
• Improving healthcare – Data-driven medicine involves analysing vast numbers of medical
records and images for patterns that can help spot disease early and develop new
medicines.
• Big Data could immensely aid in the management and analysis of observational data.
• Combining both social media postings and clinical data can uncover behavioural patterns
that are otherwise hard to detect by other means.
Using Social Media (Twitter) To Predict Depression
• Social media provides an important space for many people dealing with rare and often
stigmatized mental illnesses.
• Many individuals turn to Facebook groups or online gaming platforms to connect with others
who share similar experiences.
• What do twitter profiles and posted images reveal about depression and anxiety?
• Ungar et al., 2019 found that the colour, lighting and symmetry of photos shared on Twitter
may give insight into who is at risk for anxiety and depression.
• Those that had depression were less likely to have a profile picture that was happy or that
exhibited a particular mood
• Demonstrates the use of pictures as data.
• https://arxiv.org/pdf/1904.02670.pdf
Using big data to test whether a man and women will (unite) go on a second date
Can you predict whether a man and women will go on a second date based on how they speak on
the first date?
• The authors used traditional data (height, weight, hobbies), together with recordings of the
conversations.
• The scientists coded the words used, the presence of laughter and tone of voice.
