Machines may not have human emotions but over the last few decades they have evolved into not just smarter computing devices, but have also become good at reading emotional reactions of humans. Now, artificial intelligence is helping autistic children learn to interact with others and reach their full potential. AI is also helping fight emotional problems such as anxiety and depression.
Using AI To Detect Autism
Now, a study has suggested that with the help of AI, children with autism spectrum disorder (a developmental disorder that affects communication and behaviour) can improve their communication and social skills by almost 30 percent. AI and robots are great tutors for autistic children because they can repeat instructions, stories and answers numerous times, without getting fatigued. Children, in turn, are able to improve their social skills and maintain relationships with the AI-based device. In fact, children with autism are more open to robots and AI than they are to humans. Researchers have found out that it’s because robots are not as complex as humans when it comes to emotional communication; they are kind of like toys.
Many organisations are drawing a framework for developers to understand the potential risks associated with machine learning applications and how to combat marginalisation and discrimination of humans in AI.
AI Can Detect Autism At Early Stage
AI can even detect the changes in brain function of six-month-olds and predict if the children would later develop autism.
Researchers from the University of North Carolina applied machine learning to detect changes in brain function of six-month-old babies to find out if they would later develop autism. ML algorithms analysed how the activity of each piece of the brain was synchronised with another piece of the brain. Interestingly, the algorithm was able to pick out functional connections which corresponded with autistic tendencies. According to reports, the ML algorithm was able to predict that nine out of the 11 kids would go onto be diagnosed with autism. It turned out to be true after a round of medical checkup. Early detection gives autism neuroscience a big leg up and helps the doctors as well as the family to implement changes to better manage their condition.
How Humanoids And Apps Are Helping Autistic Persons
There are many robots that are helping persons with autism in the evaluation, therapy and communication skills. For instance, a humanoid robot, Nao learns about a child’s behaviour by using two cameras and four microphones to record the child’s facial expression and body language as it interacts. Once it records the data, it carefully assesses the data to figure out the most effective way to gain the child’s attention. However, it is still in its testing stage and is expected to become available in 2019.
Robokind’s Milo that shows emotions through facial expressions and can communicate with its own voice, can teach children about social norms. Through applied behaviour, this humanoid robot provides immediate feedback, so that the kids know whether their responses are right or wrong.
Identifor Companion app is helping adults and children find employment. It includes an AI-powered virtual assistant called Abby to have real back-and-forth conversations with the users. It learns the routines of users and keeps their school, work and social life on track.
Even virtual assistants like Siri, Cortana and Alexa which aren’t designed to cater to these people, have proven to have the right skill sets to communicate with them. Judith Newman, author of the noted New York Times op-ed column, also wrote a book about her son’s frequent chats with Siri.
Using AI To Fight Emotional Problems
The state of mental healthcare in India is in crisis. According to a report, nearly 57 million Indians are affected by depression and 36 percent are likely to suffer a mental health ailment at some point in their lives. India is one of the most depressed countries in the world.
In a country where people are averse to talking about mental and behavioural illness such as depression, a chatbot by this startup is proving to be revolutionary by aiding those suffering from it. WYSA by Touchkin is an emotionally intelligent chatbot that acts as a virtual coach, helping in managing and improving mental health, and is claimed to be the first of its kind by the company. Wysa, an AI-enabled coach touch bases of this serious concern at hand by providing AI enabled coach and chat platform.
Conversational chatbots can deliver psychological intervention based for anxiety and depression based on cognitive behavioural therapy. They are quite user-friendly and can tailor interventions to a person’s emotional state and clinical needs. For instance Woebot, offers responses appropriate to patients’ stated mood and also sues a decision-tree type model to get to know its chat buddies over time and offer personalised responses. Its automated caring conversational style is modeled on empathy and provides weekly charts depicting their mood patterns.
Artificial intelligence can also help detect suicides. Vanderbilt University researchers designed an AI model to predict suicide risk, using electronic health records. Apps like mind.me is using machine learning to help diagnose, manage and predict mental illness. It is a passive mobile app that requires no input or effort from the users and when it detects depression, it shows the user a video message from their therapist, give them emergency contact numbers or suggest playing a phone game. It also lets users log triggers and emotions to help their therapist track their progress.
With the easy access to latest technologies like chatbots and now humanoids, it will give a much-needed respite to the woes of persons who suffer from autism or other disorders which are stigmatised in the society. Even though AI and machines lack emotions, they are making people feel a lot more comfortable in sharing their true feelings, knowing that the machine will help them and not judge them. However, the emergence of such AI-based alternatives raises another question — will the conventional methods of emotional and mental health care that involve human interaction become obsolete?
Try deep learning using MATLAB