1. BEHAVIORAL DISORDERS
SCREENING AND
PREDICTION
By
Sonu Kumar
Head of AI CoE
Orbit Shifters Inc.

2. INTRODUCTION
2
1

3. “Behavioral disorders, also known as disruptive
behavioral disorders, are the most common
reasons that parents are told to take their kids for
mental health assessments and treatment.
Behavioral disorders are also common in adults.
If left untreated in childhood, these disorders can
negatively affect a person’s ability to hold a job
and maintain relationships.
33

4. In the United States, 11% to 20% of children have
a mental or behavioral disorder. However, only
an estimated one in eight children receives
treatment. (Survey report in 2015)
4

5. WHAT ARE THE TYPES
OF BEHAVIORAL
DISORDERS?
5
2

6. According to multiple reports, behavioral disorders may be
broken down into a few types, which include:
▰ Anxiety disorders
▰ Disruptive behavioral disorders
▰ Dissociative disorders
▰ Emotional disorders
▰ Pervasive developmental disorders
6

7. ATTENTION DEFICIT HYPERACTIVITY DISORDER
(ADHD)
According to Centers for Disease Control and Prevention, ADHD
is a condition that impairs an individual’s ability to properly focus
and to control impulsive behaviors, or it may make the person
overactive.
ADHD is more common in boys than it is in girls. According to
the Wexner Medical Center at Ohio State University, males are
two to three times more likely than females to get ADHD.
7

8. EMOTIONAL SYMPTOMS
OF BEHAVIORAL
DISORDERS
8
3

9. 9
According to Boston Children’s Hospital, some of the emotional
symptoms of behavioral disorders include:
▰ Easily getting annoyed or nervous
▰ Often appearing angry
▰ Putting blame on others
▰ Refusing to follow rules or questioning authority
▰ Arguing and throwing temper tantrums
▰ Having difficulty in handling frustration

10. PHYSICAL SYMPTOMS
OF BEHAVIORAL
DISORDERS
10
4

11. 11
Unlike other types of health issues, a behavioral disorder will
have mostly emotional symptoms, with physical symptoms such
as a fever, rash, or headache being absent. However, sometimes
people suffering from a behavioral disorder will develop a
substance abuse problem, which could show physical symptoms
such as burnt fingertips, shaking or bloodshot eyes.

12. PREDICTION OF AUTISM:
A BEHAVIORAL DISEASE
12
5

13. 13
Autism, or autism spectrum
disorder (ASD), refers to a
broad range of conditions
characterized by challenges
with social skills, repetitive
behaviors, speech and
nonverbal communication.

14. “The prevalence of autism in the United States has
risen steadily since researchers first began
tracking it 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.
1414

15. SIGNS AND SYMPTOMS
OF ASD
15
6

16. “People with ASD have difficulty with social
communication and interaction, restricted
interests, and repetitive behaviors. The list below
gives some examples of the types of behaviors
that are seen in people diagnosed with ASD. Not
all people with ASD will show all behaviors, but
most will show several.
1616

17. 17
Social communication / interaction behaviors may include:
▰ Making little or inconsistent eye contact
▰ Tending not to look at or listen to people
▰ Rarely sharing enjoyment of objects or activities by pointing or showing
things to others
▰ Failing to, or being slow to, respond to someone calling their name or to
other verbal attempts to gain attention
▰ Having difficulties with the back and forth of conversation

18. 18
▰ Often talking at length about a favorite subject without noticing that
others are not interested or without giving others a chance to respond
▰ Having facial expressions, movements, and gestures that do not match
what is being said
▰ Having an unusual tone of voice that may sound sing-song or flat and
robot-like
▰ Having trouble understanding another person’s point of view or being
unable to predict or understand other people’s actions

19. 19
Restrictive / repetitive behaviors may include:
▰ Repeating certain behaviors or having unusual behaviors. For example,
repeating words or phrases, a behavior called echolalia
▰ Having a lasting intense interest in certain topics, such as numbers,
details, or facts
▰ Having overly focused interests, such as with moving objects or parts of
objects
▰ Getting upset by slight changes in a routine
▰ Being more or less sensitive than other people to sensory input, such as
light, noise, clothing, or temperature

20. 20
People with ASD may also experience sleep problems and
irritability. Although people with ASD experience many
challenges, they may also have many strengths, including:
▰ Being able to learn things in detail and remember information for long
periods of time
▰ Being strong visual and auditory learners
▰ Excelling in math, science, music, or art

21. CAUSES AND RISK
FACTORS
21
7

22. 22
While scientists don’t know the exact causes of ASD, research
suggests that genes can act together with influences from the
environment to affect development in ways that lead to ASD.
Although scientists are still trying to understand why some people
develop ASD and others don’t, some risk factors include:
▰ Having a sibling with ASD
▰ Having older parents
▰ Having certain genetic conditions—people with conditions such as Down
syndrome, fragile X syndrome, and Rett syndrome are more likely than
others to have ASD
▰ Very low birth weight

23. HOW DO CLINICIANS
DIAGNOSE AUTISM?
23
8

24. 24
There is no blood test, brain scan or any other objective test that
can diagnose autism—although researchers are actively trying to
develop such tests. Clinicians rely on observations of a person’s
behavior to diagnose the condition.
In the U.S., the criteria for diagnosing autism are laid out in the
“Diagnostic and Statistical Manual of Mental Disorders” (DSM).
The criteria are problems with social communication and
interactions, and restricted interests or repetitive behaviors. Both
of these ‘core’ features must be present in early development.

25. WHAT IS THE
PREVALENCE OF AUTISM
IN THE U.S.?
25
9

26. 26
The Centers for Disease Control and Prevention (CDC) estimates that 1 in 68
children in the U.S. have autism. The prevalence is 1 in 42 for boys and 1 in
189 for girls. These rates yield a gender ratio of about five boys for every girl.
How Artificial Intelligence can help in prediction and prevention?
1. An early diagnosis of neurodevelopmental disorders can improve
treatment significantly and decrease associated healthcare costs.
2. Artificial intelligence can predict whether a person is suffering from
autism or not!

27. OUR WORK
27
10

28. 28
Autistic Spectrum Disorder (ASD) is a neurodevelopment condition associated with significant healthcare
costs, and early diagnosis can significantly reduce these. Unfortunately, waiting times for an ASD diagnosis are
lengthy and procedures are not cost effective. The economic impact of autism and the increase in the number
of ASD cases across the world reveals an urgent need for the development of easily implemented and effective
screening methods. Therefore, a time-efficient and accessible ASD screening is imminent to help health
professionals and inform individuals whether they should pursue formal clinical diagnosis. The rapid growth in
the number of ASD cases worldwide necessitates datasets related to behavior traits. However, such datasets
are rare making it difficult to perform thorough analyses to improve the efficiency, sensitivity, specificity and
predictive accuracy of the ASD screening process. Presently, very limited autism datasets associated with
clinical or screening are available and most of them are genetic in nature. Hence, we propose a new dataset
related to autism screening of adults that contained 20 features to be utilized for further analysis especially in
determining influential autistic traits and improving the classification of ASD cases. In this dataset, we record
ten behavioral features (AQ-10-Child) plus ten individuals characteristics that have proved to be effective in
detecting the ASD cases from controls in behavior science.

29. DATA ATTRIBUTES
29
11

30. 30
Attribute Type Description
Age Number years
Gender String Male or Female
Ethnicity String List of common ethnicities in text format
Born with jaundice Boolean (yes or no) Whether the case was born with jaundice
Family member with PDD Boolean (yes or no) Whether any immediate family member has a PDD
Who is completing the test String Parent, self, caregiver, medical staff, clinician ,etc.
Country of residence String List of countries in text format
Used the screening app before Boolean (yes or no) Whether the user has used a screening app
Screening Method Type Integer (0,1,2,3) The type of screening methods chosen based on age category (0=toddler, 1=child,
2= adolescent, 3= adult)
Question 1 Answer Binary (0, 1) The answer code of the question based on the screening method used
Question 2 Answer Binary (0, 1) The answer code of the question based on the screening method used
Question 3 Answer Binary (0, 1) The answer code of the question based on the screening method used
Question 4 Answer Binary (0, 1) The answer code of the question based on the screening method used
Question 5 Answer Binary (0, 1) The answer code of the question based on the screening method used
Question 6 Answer Binary (0, 1) The answer code of the question based on the screening method used
Question 7 Answer Binary (0, 1) The answer code of the question based on the screening method used
Question 8 Answer Binary (0, 1) The answer code of the question based on the screening method used
Question 9 Answer Binary (0, 1) The answer code of the question based on the screening method used
Question 10 Answer Binary (0, 1) The answer code of the question based on the screening method used
Screening Score Integer The final score obtained based on the scoring algorithm of the screening method
used. This was computed in an automated manner

31. WHAT ARE THE STEPS ?
31
12

32. 32
1. ASD screening using machine learning
2. Introducing the dataset
3. Splitting the dataset into training and testing datasets
4. Building the network
5. Testing the network

33. DATA FOR THE FIRST
PATIENT
33
13

34. 34
The patient does not have
autism is a six-year-old child
who lives in Jordan.

35. CORRELATION BETWEEN
THE QUESTIONS
35
14

36. 36
We will see the correlation
between all the 10
questions:

37. OUR MODEL’S
ACCURACY
37
15

38. 38
1. We get an accuracy of 1
which means 100% on
the training dataset

39. 39
2. Testing our model on
unseen data we get
accuracy around 95%

40. 40
We find that precision is going to be a false positive while recall
takes into account false negatives. The f1-score is a combined
score of those precision and recall scores. The support value
shows the total number of patients that did or didn't have autism
in our testing dataset. We had 30 patients that had autism and
29 who didn't. This shows that the precision of our autism class
is accurate, which means that, if a patient does have autism,
we're likely to predict it. Similarly, we had very few false
positives, which also means that we're very likely to predict when
somebody doesn't have autism.

41. CONCLUSION AND
FUTURE WORK
41
16

42. 42
1. Thus, to summarize, we set out with the hopes of applying machine learning
algorithms, specially, supervised machine learning techniques that can
classify new patients (new instances) with certain measurable characteristics
(the variables) into one of two categories patient has ASD" or patient does
not have ASD".
2. In our consideration, to build an accurate and robust model, one needs to have
larger datasets. Here the number of instances after cleaning the data was not
sufficient enough to claim that this model is optimum. Looking at the
performances of our learning models, nothing can be improved with this
current data set as models are already at their best. After discussing this issue
with a researcher directly working on adult autism, we have realized that it is
extremely difficult to collect a lot of well documented data related to ISD.

43. 43
THANKS!