The use of natural language processing to make progress in mental health research, discussed in several research papers, has seen some success. Combining big data and machine learning can enhance artificial intelligence and lead to even more advancements in mental health diagnoses and solutions.
IBM, for example, has been conducting research in how cognitive computers can analyse a patient’s speech or written words and combine results from this analysis to those gained from wearable devices and imaging systems (MRIs and EEGs) to paint a complete picture of the individual’s health and better treat their health. By analysing speech or written words for early detection of schizophrenia or depression and to evaluate our mental health, cognitive assistants and sensors in our smartphones (with the support of automated mental health tools) can help advance mental health support.
Startups such as Mindstrong are using AI to turn smartphones into an emotional diagnostic device. By tracking an individual’s physical interactions with their phone, that information can be used to form a picture of their mental state. While psychiatrists may not pick up on signals presented by patients, such as the words they use and the tone of their voice – all of which have been proven to have a correlation with mental health issues – AI can become more reliable and useful as it will be able to notice these signs. For example, a research team recently developed a workable AI model that predicted which members of an “at-risk” group of young people would develop psychosis within a time frame and which would not.
An even more interesting solution emerging from the use of big data and AI to provide mental health solutions would be the development of an AI app or bot that can empathise. While at the moment there are possibilities for AI to help people with their emotions, there is no emotionally intelligent AI. Programming a machine to learn how and when to display emotions, rather than appear as if they have emotions, will help those struggling with their mental health because they will be provided with the wealth of knowledge the machine can offer as well as emotional support.
Ethically, however, this would be problematic and would require support from patients, clinicians and researchers. The potential advantages using big data for the advancement of mental health research and solutions are significant; nevertheless, none of this – the data collection and the approaches to the research – is possible without a discussion on ethics, privacy and security.
Written by Rodney
If interested in the references, or would like more information, contact us via email: email@example.com