Recently, there has been a lot discussion around artificial intelligence (AI) and machine learning in the diagnosis and treatment of autism and other similar disorders.
There is clearly vast potential, but the critical point is that these technologies must both learn from and serve existing medical trends and protocols. The fundamental nature of child development and social communication is not well understood by human experts, so technology must aid, not oversimplify, its approach to symptoms, conditions and treatment associated with autism, Fragile X or other similar disorders.
For machine learning to truly be effective, it requires vast amounts of training data. With the advent of connected/IoT devices in the marketplace, we will see an increase in the frequency and variety of the types of signals that can be automatically and passively captured, providing a slew of real-time data for machine learning algorithms to learn from.
Trayt is well positioned to contribute to this trend. We are working to gather information from sleep monitors, heart rate sensors, audio sensors, connected cameras and environmental sensors; then, we will correlate it with additional information captured from the Trayt app, such as symptoms severity and frequency, genetic information, medication and therapy adherence information, side effects, patient demographics, medical background and more.
We will not only be able to continuously expand the breadth of information available from what we capture, but also offer a much more nuanced set of insights to our users. This ongoing addition of sophisticated data will help to refine the findings and allow for a real-time view of progress of patient outcomes, which will in turn help to tailor treatment and optimize the effectiveness of medications and therapies.
Robots are also being considered as a tool for treating autism symptoms. However, my view is that for kids with social-communicative issues, nothing can replace face-to-face interactions with other kids, their therapists or other adults guiding the interactions. Robots have a place as a prop in the environment where they can capture signals - like visual posture, face emotions and sound to capture voice - which can then be fed into machine learning algorithms to measure the effectiveness of the therapies and patient outcomes over time. However, they should not be used in place of real human interaction.
While there has been positive progress in the application of advanced technology towards treating autism and similar disorders, it should continue to do so under the caution that it is not, and might not ever be, in a place where it can be relied upon independently. The human mind is far too complex to currently be captured and adequately transcribed into a line of code - especially the minds of children who are still in their developmental phases and are affected by social communication disorders.
Trayt helps bridge the connection between authentic human interactions and the largescale sourcing of data that will help to inform these technologies and continue to improve the lives of patients with autism and other disorders.