Machine Learning is Developing to Augment Diagnosis and Treatment of Autism (ASD)

A.I is Advancing at the Intersection with Autism

If you enjoy articles about A.I. at the intersection of breaking news join AiSupremacy?here. I cannot continue to write without community support. (follow the link below). Join 77 other paying subscribers.

https://aisupremacy.substack.com/subscribe

Not just early diagnosis

Image: Photo by Emily Wade on Unsplash

Hey Guys,

I’d like to discuss not just Autism and AI, but being on the spectrum.

How many people have autism? According to the Centers for Disease Control and Prevention (CDC), about 1% of the world's population has autism spectrum disorder –?over 75,000,000 people. That may be a large number, but autism spectrum disorder (ASD) features a wide range of symptoms and levels of severity.

I’ve often been fascinated by how ASD can go undiagnosed by how people, often women, are skilled at masking it. Many adult women therefore do not know they are on the ASD spectrum. There have been many articles written about why this is, for instance the Conversation back in 2017 wrote:

Because such women are intelligent and used to compensating for communication impediments they may not be consciously aware of, they can slip through the cracks of our still-inefficient diagnostic procedures.

I know how this might work as I grew up with a speed impediment and how it shaped me socially and in terms of avoiding certain kinds of communication to try to protect myself.

Fast forward to 2022, when the early diagnosis of ASD is considered crucial, Artificial intelligence (AI) augments a doctor’s ability to diagnose, treat, and understand disease. Medical experts all around the world use AI to address common conditions.

There’s more research, case studies and applications now reaching the field and showing great promise.

Think about it, the one percent with ASD face special challenges. Picture your child potentially is showing early symptoms.?Autism spectrum disorder?(ASD) is a challenge in multiple ways. Just getting diagnosed can take months of visits to doctors and specialists. After the diagnosis, children are often put on long waiting lists to begin therapy, which itself consists of frequent sessions that while helpful, are usually quite taxing for both the children and their parents.

AI detects autism speech patterns across different languages

A new study at?Northwestern, really caught my attention. What impressed me about it is the researchers believe their work could provide a tool that might one day transcend cultures, because of the computer’s ability to analyze words and sounds in a quantitative way regardless of language.

Just as A.I. is perfecting Universal language translation, A.I. is soon going to be able to understand the language of ASD. A new study led by Northwestern University researchers used machine learning — a branch of artificial intelligence — to identify speech patterns in children with autism that were consistent between English and Cantonese, suggesting that features of speech might be a useful tool for diagnosing the condition.

Applied Medical Devices

In primary care settings there’s also new diagnosis tools with A.I. and medical devices. New research has found that an artificial intelligence (AI)-based medical device can assist clinicians in primary care settings accurately diagnose autism spectrum disorder (ASD) in children up to 6 years old.

The?study, published in?npj Digital Medicine?earlier this month, notes that ASD can be reliably diagnosed as early as 18 months, but diagnostic delays persist in the US. ASD is also one of the most common developmental disorders among children, and early intervention has been shown to improve long-term outcomes.

Clearly A.I. is on track to improve early diagnosis of ASD and make millions of lives just a little bit easier. For people who have lived with ASD of different severities there’s a realization of how life may be different for the next generation of people with these kinds of brains.

In society at large, just as the stigma of having mental health issues has lessoned so has the stigma around being “neuro-divergent”. A.I. is normalizing a new generation of how various issues are diagnosed and recognized as not as fringe after all. Being “on the spectrum” is no longer taboo.

All humans follow patterns that A.I. can notice better than any human clinician. The NorthWestern study was undertaken with collaborators in Hong Kong, the study yielded insights that could help scientists distinguish between genetic and environmental factors shaping the communication abilities of people with autism, potentially helping them learn more about the origin of the condition and develop new therapies.

But could A.I. also help design and implement some of the therapies and treatments? Researchers are exploring the idea that artificial intelligence (AI) could be used to diagnose autism and help people on the autism spectrum to improve social, communication, and emotional skills.

A new generation of consumer Apps is trying to tackle this issue. Diagnosis of autism through the use of AI is now a reality and AI-based therapies in development show promise.?Some AI-based apps are now downloadable?for any smartphone user. The best of those may evolve into more personalized applications.

Language and Autism and A.I.

Children with autism often talk more slowly than typically developing children, and exhibit other differences in pitch, intonation and rhythm. But those differences (called “prosodic differences'' by researchers) have been surprisingly difficult to characterize in a consistent, objective way, and their origins have remained unclear for decades.

However, a team of researchers led by Northwestern scientists?Molly Losh?and Joseph C.Y. Lau, along with Hong Kong-based collaborator Patrick Wong and his team, successfully used supervised machine learning to identify speech differences associated with autism.

See Molly here: “Genes and Environment Interact in Autism” in 2017.

Why Not Read Me on the App?

Read AI Supremacy in the new Substack app

Now available for iOS

Get the app

The results were?published in the journal?PLO One?on June 8, 2022.

As someone who studied psychology, I found the study pretty fascinating.

“When you have languages that are so structurally different, any similarities in speech patterns seen in autism across both languages are likely to be traits that are strongly influenced by the genetic liability to autism,” said Losh, who is the Jo Ann G. and Peter F. Dolle Professor of Learning Disabilities at Northwestern. “But just as interesting is the variability we observed, which may point to features of speech that are more malleable, and potentially good targets for intervention.”

We don’t fully know the extent to which A.I. will be able to personalized education, healthcare and even therapies in the future to the individual, but I think it’s might become pretty significant. The treatment of ASD has potentially a special relationship with artificial intelligence.

There’s also unique startups in this space. SpectrumAi, founded by a mother of four children with autism, aims to use technology to improve therapies for children with the neurological disorder. Fast company recently profile them?here.

Ling Shao, who is the mother of four children with autism spectrum disorder, is a health industry veteran. She has held executive leadership roles at insurer UnitedHealth, healthcare services giant Optum, and digital healthcare startup Buoy Health.

Painpoint of current System

But she grew increasingly frustrated with what she sees as some of the biggest barriers to more widely accessible autism care: a lack of data transparency that makes defining—and, consequently, seeking out—proven, high-quality care difficult for parents.

Can A.I. Improve it?

Last year Shao launched?SpectrumAi?to tackle some of these fundamental shortcomings in autism care, using a software platform and artificial intelligence to capture autism treatment data with an eye toward a future in which care providers are paid for the value they bring rather than the bulk of services they provide, and encouraging best practices.

Is Autism becoming More Common?

There is also some sense that Autism or its diagnosis is becoming more common. I’m not an expert but the data is startling. The latest estimate of autism prevalence—1 in 68—is?up 30 percent from the 1 in 88 rate reported in 2008, and more than double the 1 in 150 rate in 2000.

The rise in the rate has sparked fears of an autism ‘epidemic.’ But experts say the bulk of the increase stems from a growing awareness of autism and changes to the condition’s diagnostic criteria (Scientific America).

A.I. can Solve English Diagnosis Bias

Joseph C.Y. Lau in the NorthWestern study added that the use of machine learning to identify the key elements of speech that were predictive of autism represented a significant step forward for researchers,?who have been limited by English language bias in autism research?and humans’ subjectivity when it came to classifying speech differences between people with autism and those without.

Machine learning is really a game-changer in creating a more representative early diagnosis tool across languages and cultures. I like this more inclusive approach enabled by A.I. very much.

“Using this method, we were able to identify features of speech that can predict the diagnosis of autism,” said Lau, a postdoctoral researcher working with Losh in the Roxelyn and Richard Pepper Department of Communication Sciences and Disorders at Northwestern. “The most prominent of those features is rhythm. We’re hopeful that this study can be the foundation for future work on autism that leverages machine learning.”

Since A.I. can spot patterns easily that humans might miss one can witness an explosion of A.I. trying to help people and the parents of kids with ASD. I think there’s considerable progress in this field that isn’t being documented very well and a body of research that has yet to hit a publication or peer-review state. Just as new treatments for Cancer might show a lot of promise in the 2020s, I’m pretty optimistic about A.I.’s impact on ASD as I review the literature.

However our the stability of our healthcare systems and their labor pools is also very relevant. For instance, the researchers further noted?that diagnostic delay in the US is due, in part, to the increase in demand for ASD evaluations, which has outpaced specialist capacity and led to prolonged wait times. ASD diagnosis in the US also relies on a limited number of pediatric subspecialists and time-intensive, team-based behavioral evaluations, which can take as long as 18 months from initial screening to final diagnosis.

A.I. could significantly help reduce wait times?and time to treatment that leads to long-term better outcomes for ASD individuals. With new medical devices, this might also make sense at scale. The use of diagnostic aids in primary care settings to augment ASD care has shown significant potential in?previous studies, and the researchers aimed to test the accuracy of one such tool in one of the studies I read. The tool, an AI-based software-as-a-medical-device (SaMD), offers recommendations for the primary care provider after analyzing behavioral features from a caregiver questionnaire, a healthcare provider questionnaire, and two short home videos.

Just as Zoom is getting better at analyzing sales meetings, and HR now uses some AI-screening in the interview process, so too are early child videos helping to diagnose autism with the help of machine learning. I find this all pretty incredible!

Anyways guys while my coverage has not been in-depth here, I do hope this gives you something to think about regarding the positive impacts of A.I. on healthcare and ASD that impacts so many individual, families, parents and our healthcare systems around the world. We are all human, and language and A.I. is becoming more embedded, which means in a sense our society is integrating A.I. into its institutions, corporations and how we make decisions that can be more impactful all-around with better data.

Thanks for reading! Join 77 paid subscribers?if you think supporting me so that I can continue to write is something you want to do. I host a range of Newsletters on Substack and have been creating nearly every day during the last seven months. I’m very happy doing this for the foreseeable future, if I can make the economics work to support my family.

I am grateful to be able to cover these stories as I actively think about the future.

Subscribe now

I’d also like to recommend other A.I. Newsletter on Substack that each offer tremendous value.

TheSequence

The best source to stay up-to-date with the developments in the machine learning, artificial intelligence, and data science world. Trusted by 124,573 professionals from the main AI labs, universities, and enterprises

By Ksenia Semenova

Last Week in AI

weekly text and audio summaries of the most interesting AI news, as well as editorials commenting on recent events

Altdeep.ai Newsletter

A semi-weekly newsletter to hone your skills in empirical modeling and statistical thinking.

By Robert Osazuwa Ness

Including so many others. Have a good week!

If you enjoy articles about A.I. at the intersection of breaking news join AiSupremacy?here. I cannot continue to write without community support. (follow the link below). Join 77 other paying subscribers.

https://aisupremacy.substack.com/subscribe