Photo Source: Good Free Photos
Feelings of despondency and loss of interest in things enjoyed in the past can indicate that a person has depression. The National Institute of Mental Health describes depression as, “…a common but serious mood disorder. It causes severe symptoms that affect how you feel, think, and handle daily activities, such as sleeping, eating, or working. To be diagnosed with depression, the symptoms must be present for at least two weeks.” (nimh.nih.gov/health) Depression also stands out as one of the most significant contributors to the global burden of disease. Health researchers use a measurement known as Disability Adjusted Life Years, or DALY for short, to compare the relative impact of different diseases on the human population of the world. One DALY represents a year of being sick instead of healthy. In a study performed by AJ Ferrari at the University of Queensland, Australia and others titled “Burden of Depressive Disorders by Country, Sex, Age, and Year: Findings from the Global Burden of Disease Study 2010” published in PLoS, the authors determined from a statistical analysis of health records from around the world using DALYs that depression occurs more commonly in females and adults of working age, but the study also noted measurable depression in children in the 5-9 year-old category that increases through adolescence and peaks in the 20-24 year-old age range. Children for example may display depression in what psychologist call anhedonia, which means loss of pleasure and enjoyment of things they liked before.
Currently, no specific physical test exists that determines if a patient has depression. Unlike other areas of medicine where blood tests, x-rays, or MRI scans help doctors to make a diagnosis of health issues such as cancer, broken bones, or tissue damage, depression requires talking. Doctors need to ask questions about people’s lives to figure out whether the patient has depression. However, before a doctor even gets involved, according to Anxiety and Depression Association of America, only 36.9% of people suffering from depression get treatment and many more go undiagnosed.
Advances in artificial intelligence in areas of speech and facial recognition may help healthcare providers better diagnose depression and even help individuals with self-diagnosis. Remarkable research at MIT recently demonstrated that a form of artificial intelligence known as deep learning can detect depression by listening to a person’s conversation or analyzing written answers to questions. The work by Tuka Alhanai, Mohammad Ghassemi, and James Glass at the Computer Science and Artificial Intelligence Laboratory in Cambridge, MA, presented at a conference called Interspeach 2018 and summarized in “Detecting Depression with Audio/Text Sequence Modeling of Interviews,” the authors describe their experiments. They concluded that a depressed person will speak and write differently than a normal person and that their artificial intelligence can detect depression from their words. Interestingly, their system could detect depression from the written answers with fewer words than from speech, but the best predictions came with a combination of both written and vocal information. Such a system could passively detect the onset of depression just following people’s conversations on the phone or in the home. Another type of artificial intelligence that shows promise in detecting depression appears in the area of affective computing. Affective computing uses artificial intelligence to predict depression from changes in facial expression. In “Social Risk and Depression: Evidence from Manual and Automatic Facial Expression Analysis,” JM Girard and others demonstrated that automatic facial expression analysis can detect depression and suggest that doctors could use such a tool in the clinic.
Depression ranks as one of the top debilitating health issues facing people today across the world. With less than 37% of people getting treated for depression in the US and many going undiagnosed, better and more accessible depression diagnoses would greatly benefit people suffering from this affliction. Social stigma, awareness, cost, and convenience explains some of the reasons people do not get diagnosed with depression or subsequently seek treatment. Breakthroughs in artificial intelligence have opened the door for automated detection of depression through human conversation or facial expressions. With the escalating cost of healthcare and poor track record for depression treatment, artificial intelligence working through our smartphones or passive devices such as Amazon’s Echo or Microsoft’s Cortana to surface issues of depression that may otherwise go on for years. Depression effects quality of life, physical health, and work productivity. In the not so distant future, you can imagine a parent asking, “Alexa, do you think my kid is depressed?" and getting an answer back.
Dr. Smith’s career in scientific and information research spans the areas of bioinformatics, artificial intelligence, toxicology, and chemistry. He has published a number of peer-reviewed scientific papers. He has worked over the past seventeen years developing advanced analytics, machine learning, and knowledge management tools to enable research and support high level decision making. Tim completed his Ph.D. in Toxicology at Cornell University and a Bachelor of Science in chemistry from the University of Washington.
You can buy his book on Amazon in paperback and in kindle format here.