How can AI be used successfully to improve mental health?

How are you today? How do you feel? These simple questions, asked by most mental health therapists, might be the key to improving your life and reducing the cost of health care. But there is a catch, and it’s not the one most often cited, which is the need to use these services to gain the benefit. No, there is a darker problem, a foundational crack in the healthcare delivery model… there are not enough practitioners to treat everyone. Or is there?

Artificial Intelligence (AI) is already transforming different industries, including mental health. Data sciences, machine learning, and artificial intelligence technologies enable a new and unfamiliar type of mental health care that focuses on providing “individualized emotional support.” Using deep neural networks, an advanced AI, enable the delivery of traditional therapeutic service like cognitive-behavioral therapy (CBT) for pennies on the dollar, augmenting the traditional skills seen only in human doctors. 

Woebot, a Facebook-based application development by Dr. Alison Darcy, is one example of these digital health technologies that focus on the mind to drive changes in the body. Woebot mimicked the conversation a patient has with their therapist. It not only listens to what they say (intellect...IQ), but why it was said (emotion... EQ). Through a deeper understanding of the emotional drivers behind our rational thinking, the virtual delivery of therapeutic techniques like CBT are possible. That is an important therapeutic first step.

Mental health is now the most expensive part of our health care system, outpacing its physical health counterpart, surpassing the treatment costs of heart conditions. Mental health is a causal factor in developing many physical ailments. Schizophrenia and bipolar disorders are causal factors for many types of cardiovascular diseases. So the early detection and treatment of mental conditions can play a significant role in the reduction of many physical ailments. Leveraging AI in the detection and treatment of these conditions can provide ready access treatment to people who have been otherwise socially or financially disenfranchised.

DuPage County is one of the largest of the 2,800 local health departments across the United States. Serving over 932,000 residents, the dedicated staff of 600 health care professionals provides a wide range of mental and physical services to the increasing needs of their constituents. As excellent stewards of their county’s fixed resources, DuPage is researching the benefits of artificial intelligence for their mental and physical health services. Karen J. Ayala, Executive Director of DuPage County Health Department said, “I believe we will be better served, and improve overall health in our communities, through engagement with these resources [AI].” Karen Ayala is at the forefront of those looking to help people through the application of AI. But not all AI is useful in this unfamiliar world of digital therapeutic technologies.

In order for AI to be a part of an effective treatment program, in order for it to help us, AI must be as personal as traditional face-to-face therapy sessions. Artificial intelligence can not rely on modern pre-baked models trained with generic examples sets. These convolutional neural networks (CNN) and recurrent neural networks (RNN) are the staples of the modern-day data scientist tools set. While they are sophisticated structures of deeply interconnected networks of simulated neural pathways, they still can not represent and understand the mental modalities that make each of us unique. Make each of us, us. This is where Evolutionary Computing can come to the proverbial rescue by OE letting AI create AI.

With evolutionary computing (EC), extremely deep and interconnected neural systems can be organically created based on personalized behavioral traits. Personalized observations are used to create random populations of behavioral models. EC test these models to see how they compare against our observed behaviors. These are personalized tests based on you and me, not some generalized group behavior, resulting is individual digital surrogates. 

Evolutionary computing combines the best of these digital surrogates, like mental DNA, to create the next generation of behaviors. This process continues, incorporating new observations over millions of evolutions, until they create an optimal personalized behavioral match. The result... a predictive behavioral model that is you... a digital you. Evolutionary computing is the next generation of AI that creates itself. EC is like baking a cake from scratch instead of from a box... you get the taste and texture that suits your needs.

The care of our health is becoming more sophisticated as the world becomes more complex. To prevent some future physical disease, your doctor might have to treat a mental condition today. Evolutionary computing based artificial intelligence is one way that we can bring therapeutic resources to those with needs that scale on the backs of computers, reducing the dependency on costly human-based skilled and otherwise unavailable professionals. So, how are you today and how do you feel? The next time they ask you these questions, you just might be talking with Dr. AI.

If you enjoyed the post, please click the thumbs up icon and let me know!

Looking For Some Sleeves Rolled Up Hands-On Support?

For those who want to discuss their specific Data Sciences and Artificial Intelligence needs, I’m happy to invest in a discovery call. I invite you to reach connect with me, follow me on LinkedIn here, and learn more about me here.