www.nature.

com/scientificreports

OPEN Detection of autism spectrum

disorder (ASD) in children
and adults using machine learning
Muhammad Shoaib Farooq 1, Rabia Tehseen 2, Maidah Sabir 1 & Zabihullah Atal 3*

Autism spectrum disorder (ASD) presents a neurological and developmental disorder that has an
impact on the social and cognitive skills of children causing repetitive behaviours, restricted interests,
communication problems and difficulty in social interaction. Early diagnosis of ASD can prevent from
its severity and prolonged effects. Federated learning (FL) is one of the most recent techniques that
can be applied for accurate ASD diagnoses in early stages or prevention of its long-term effects. In
this article, FL technique has been uniquely applied for autism detection by training two different ML
classifiers including logistic regression and support vector machine locally for classification of ASD
factors and detection of ASD in children and adults. Due to FL, results obtained from these classifiers
have been transmitted to central server where meta classifier is trained to determine which approach
is most accurate in the detection of ASD in children and adults. Four different ASD patient datasets,
each containing more than 600 records of effected children and adults have been obtained from
different repository for features extraction. The proposed model predicted ASD with 98% accuracy (in
children) and 81% accuracy (in adults).

Autism is categorized as neuro-developmental disorder which has severe effects on social growth and develop-
ment in children and adults. Although its complete cure seems not possible but early diagnosis is preferable as it
helps in more effective treatment compared to conventional behavioural investigations that take much time in
detecting and diagnosing ASD by analysing children behaviour in c­ linics1. ASD has been mostly diagnosed in
2 years old child but it can be diagnosed in children later depending on complexity of symptoms and severity of
the ­disorder2. It has generally occurred due to environmental factors or any genetic linkage which not only effects
the nervous system but also has an overall impact on social and cognitive skills of the children and adults. The
extent and the intensity of its symptoms are quite variable. Common signs of the condition include difficulty in
communication particularly in social situations, obsessional interests, and repeated m ­ annerisms3. A complete
examination is needed to detect ASD comprising thorough evaluation and series of assessments performed by
child healthcare professionals and psychologists. Early treatment and diagnosis of ASD are crucial since they
help to somewhat lessen symptoms, which enhances the person’s overall quality of l­ ife4. However, a lot of criti-
cal time can be lost in diagnosing ASD because it cannot be properly detected by depicting only behaviours of
children or adults in clinic. Autism can be identified as early as possible using a range of clinical approaches,
but actually these are time-consuming diagnostic procedures infrequently carried out unless the predictive risk
of ASD development is ­high5. Machine learning (ML) gives an opportunity to train ASD models in less time
and more ­accuracy6. ML techniques are crucial for quick and accurate assessment of ASD risk and streamlin-
­ uickly7. Various
ing the entire diagnostic process which assist families in getting to the critical therapies more q
classification models of ML can be used for early prediction of autism to prevent its prolonged effects in adults
as well as c­ hildren8.
Many other computational techniques have also been proposed in ­literature9 such as Hosseinzadeh et al.10
proposed IoT based solution for ASD detection and Eslami and ­Saeed11 presented deep learning based model
for healthcare of ASD effected patents. However, obtaining huge amount of data for model training in central-
ized or distributed environment remained a challenge. Hospitals hesitate to share their data as data are the most
valuable asset and regional data protection legislations also prohibit data ­sharing12. Data owner organizations
have many serious concerns about data privacy, data security and data protection. Moreover, transmission of big
dataset over the network for training machine learning model introduces further barriers of network latency,

1
Department of Artificial Intelligence, University of Management and Technology, Lahore 54000,
Pakistan. 2Department of Computer Science, University of Central Punjab, Lahore 54000, Pakistan. 3Department
of Computer Science, Kardan University, Kabul 1007, Afghanistan. *email: [email protected]

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 1

Vol.:(0123456789)
 www.nature.com/scientificreports/

communication delay and data ­theft13. Therefore, it is the immense need of time that a model should be proposed
in which data remain safe with owner organization.
Federated Learning (FL) technique is the most advanced approach of ML in which data remains secure with
owner organization and small sized local ML based classifier is trained onsite without moving data over the
­network14. FL is very beneficent in ensuring data security as data are not being shared over the network therefore
data privacy, data protection and data security issues are automatically ­resolved15. Moreover, network issues will
not be raised as only small sized local data model is travelling over the network towards central server instead of
huge ­data16. Many researchers have applied FL for detection of multiple neurological ­disorders17. Ali et al.18 have
applied FL for the detection of colon cancer using pixel level segmentation dataset. Ghosh et al.19 have applied
FL for medical image segmentation. Nigmatullina et al.17 proposed a digital platform to monitor and support
children with ASD using FL. Novelty of our work is the application of FL technique for detection of ASD in both
children and adults. Two different ML models including SVM and LR have been trained locally using four differ-
ent ASD datasets of features containing records about children and adults obtained from free sources and data
providing agencies listed in Table 1 for autism detection. We have also compared the results of proposed model
with already proposed ASD detection methods and comparable accuracy has been obtained. Major contribution
of this work is the combination of different local ML based models for training central FL based meta classifier
on features dataset of children and adults to detect ASD risk factors with reasonable accuracy.
Our article is organized in multiple sections. In “Introduction” section presents introduction of the autism
detection approaches. Most recent studies conducted on autism detection have been summarized in “Related
work” section. Research methodology, experimentation, analysis and results have been presented in “Material
and method” section. Results have been discussed in “Discussion” section. Conclusion and future directions
have been illustrated in “Conclusion” section.

Related work
Autism spectrum disorder (ASD) is a neuro-developmental disorder that results various impairments in social
interaction, communication, and the existence of unvaried patterns of behaviour in children and ­adults20.
­Alfalasi21 reported that in United States 1 out of 54 children is affected by autism. Detecting autism earlier in
one life can make a big difference than treating it l­ ater22. According to World Health Organization (WHO) every
year one among 160 children is diagnosed with ASD traits all over the w ­ orld23. Treating ASD earlier is always
24
the best option for toddlers as they are still ­developing .
Different symptoms identified in ASD patients have been considered as features that can be used for ASD
detection. Lawan et al.25 and Cantin-Garside et al.26 observed behavioural disorder, Beary et al.27 and Derbali
et al.28 recorded facial expression disorder and Devika et al.29 observed structural disorder in ASD effected per-
sons. Emotional disorder in ASD affected persons has been studied by Makhnytkina et al.30 and mental disorder
has been analysed in Liu et al.31 and Lord et al.32. Many researchers explored medical imageries for ASD detection
including Bilic et al.33, Husna et al.34, Liu et al.35, Nogay and ­Adeli36. Images of brain have been used by Subah
et al.37, Xu et al.38, Yin et al.39, Shenouda et al.40 to detect ASD in patients. Single and cross order strategy for ASD
detection has been proposed in Wawer et al.41.
Researchers have used wearable devices containing sensors for detection of ­ASD42,43. Application of intelli-
gent approaches present advanced ways to economically detect ASD effected children and ­adults44. Models have
been proposed in the literature describing application of different methods and approaches for ASD detection
like structural M ­ RI45, neural n ­ etworks46, machine l­ earning47–49, deep l­ earning50, transfer l­ earning51,52 and IoT 53.
All these techniques have been applied to detect ASD with reasonable accuracy in children and adults but faced
limitations of data acquisition as hospitals hesitate or refuse to share patient records due to organizational policies
and regional data protection legislations. Data security, data privacy and data availability are the huge challenges
in developing effective intelligent models. Even if access to data is granted, transferring huge dataset over the
network is again challenging, rising a lot of network issues regarding network congestion, latency and data theft.
Federated learning (FL) provides a generous solution to address all above mentioned problems. FL is an
advanced ML based approach that never transmits data over the n ­ etwork54. Data is kept with its generating
55
­organization whereas only a small sized local data model is trained from onsite data and transmitted over the
network towards central server where all local models are combined to train meta classifier for determining which
ML model is most effective in autism ­detection56. Objective of proposed model is to detect ASD symptoms at
different stages of age with minimum time, controlled expense and maximum accuracy. Novelty of our work is
the application of federated learning technique for autism detection in children and adults by processing four
different datasets by training SVM and LR classifiers locally. Major contribution of this work is the detection

Categories Source No. of instances No. of attributes

A: https://​archi​ve.​ics.​uci.​edu/​ml/​datas​ets/​Autis​tic+​Spect​rum+​Disor​der + Screen-
692 21
Children ing + Data + for + Children++
B: https://​www.​kaggle.​com/​datas​ets/​fabde​lja/​autism-​scree​ning-​for-​toddl​ers 654 19
C: https://​archi​ve.​ics.​uci.​edu/​ml/​datas​ets/​Autism+​Scree​ning+​Adult 704 21
Adults
D: https://​www.​kaggle.​com/​datas​ets/​andre​wmvd/​autism-​scree​ning-​on-​adults 700 19

Table 1.  Dataset description of children and adults.

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 2

Vol:.(1234567890)
www.nature.com/scientificreports/

of ASD by the application of most advanced Federated Learning technique by training ML classifiers locally on
features dataset of children and adults to find the predictive risk factors of Autism with reasonable accuracy.

Material and method

ASD indicates a disability in human development due to variations of neurons present in human b ­ rain57. Prac-
58
titioners believe that there are multivariate sources that work jointly to cause A
­ SD . Diagnosis of ASD is also
very challenging task as no medical test like blood test exists to detect ASD. Doctors usually apply psychological
and observational strategies to sense ASD in a patient by analysing multiple aspects of their daily routine as
mentioned in Fig. 1.
In this article, a unique federated learning based model has been proposed in which four different datasets of
adults and children have been analysed using LR and SVM locally to train local data models. These local models
have been transmitted towards central server for training of meta classifier in global model to predict autism in
children and adults. Proposed model architecture presented in Fig. 2 comprises of five components including
dataset acquisition, data pre-processing, ML models training for ASD detection and performance comparison
of different ML models to determine the most effective model that can accurately diagnose autism. The first step
was acquisition of data in which publicly available four datasets of children and adults from data sources listed
in Table 1 have been obtained. In second step, data pre-processing and normalization was performed for data
compression and data cleaning and removal of noisy data. After normalization, in third step, four datasets have
been locally processed by SVM and LR classifiers for autism detection. Results of training ML classifiers have
been transmitted to central server where meta classifier has been trained to compare results and identify the best

Execuve
Funcon
Verbal &
Sensory
non verbal Processin
communicao g
n
ASD
Features

Preservave Repeve
Thinking behaviors

Motor
skills

Figure 1.  Aspects observed while diagnosing ASD.

Figure 2.  Proposed model architecture.

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 3

Vol.:(0123456789)
 www.nature.com/scientificreports/

model to detect autism. In last step, results of meta classifier were validated by calculating accuracy, precision
and F1 score to detect autism disorder with more accuracy as shown in Fig. 2.

Step1: Dataset
Four datasets have been obtained covering two dimensions: children and adults. Source and specifications
of each dataset is listed in Table 1.

Step 2: Pre-processing
According to Q-Chart-10, ten different features have been unanimously identified for processing of adults and
children datasets at same scale for segregation of autism effected patients from normal ones as shown in Table 2.
The Quantitative Checklist for Autism in Children (Q-CHART-10) screening approach approved by Trans-
forming autism project, UK, served as the foundation for the conduction of this ­research3 . Thirty questions
have been asked to record responses (R1–R10) for features mentioned in Table 2. The value of these responses is
assigned to classes as per following criteria for assigning weightage (score) to every response.

Feature_found== 0
If (score >3)
{
Output: ASD feature exists
Flag == True
Feature_found == +1
}
Else
{
Flag == False
Output: ASD feature does not exist

If score of class is more than 3, it indicates that ASD feature exits, its weight is incremented by 1 and “Yes”
will be stored in response set otherwise value of flag will remain 0 that shows absence of any ASD features and
“No” will be stored in the response set. Each class variable corresponds to more than one questions confirming
the presence of feature extracted from Q-CHART-10 checklist. Information stored in class response set is in the
binary format indicating Yes (stored as 1) and No (stored as 0). Local ML models have been trained on these
responses presented in Table 2.
The response dataset contained some noisy and missing records therefore data transformations were needed
to carry out prior to train ML classifier for model training and analysis. Category variables are handled using
label encoding. To make labels machine-readable, label encoding transforms them into numeric form. Repeated
labels receive the same value as those that were previously allocated. The binary label encoding of classes with
ten features have been chosen.

Feature no Title Description Response

Patient’s age Children (0–15 years), adults (16 years or above)
Patient’s gender Male/female
1 R1
Ethnicity of patient Common ethnicities
Residence country Countries list
2 Verbal communication Response on calling Name, saying papa, mama R2
3 Non-verbal communication Eye contact, facial expressions, gestures, posture, use of objects and body language R3
4 Sensory processing See, hear, smell, taste, touch R4
5 Repetitive behaviour Do action again and again R5
6 Motor skills Walking, running, riding a bike R6
7 Preservative thinking Rumination, repetitive thinking, worry R7
8 Having jaundice Child/adult born with jaundice R8
9 Person taking ASD test Parent staff, caregiver etc. R9
10 Used the screening app before Did the user use a screening app for ASD R10
Screening method type Type of methods of screening chosen based on age category

Table 2.  Feature description.

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 4

Vol:.(1234567890)
www.nature.com/scientificreports/

Step 3: Federated Learning process

In the proposed architecture, Federated learning process starts from step three in which pre-processed and
normalized datasets have been processed for training of SVM and LR classifiers. Workflow of FL process is
presented in Fig. 3. Results of these classifiers in terms of accuracy, precision and F1 score have been calculated
and transmitted to central server for training of meta classifier at server. Meta classifier will determine which
model is more appropriate in detecting autism and will train the global model accordingly. Global model will be
disseminated in all clients as a single tool for autism detection.

Experiment. The children and adult datasets (A, C respectively) presented in Table 1 have been divided into
training and test datasets. Training datasets contained 80% records and testing datasets which will be used to test
the proposed model contained 20% of total records.

Experimental setup. Experiment has been performed in two different dimensions. In first dimension, SVM
and LR has been applied on dataset of adults presented in Table 1. In second dimension, SVM and LR has been
applied on dataset of children as presented in Fig. 4.
Results obtained after training local models have been transmitted to central server through 4G ethernet
gateway where meta classifier is trained to predict which ML model is outperforming in prediction of ASD. Best
model is selected for the training of global model that is transmitted back to the clients so that all clients use
same efficient model for autism detection.

Analysis and results. Two-dimensional exploratory analysis has been performed on datasets by plotting several
graphs to depict different perspectives of the ASD response set. In first dimension, variance between datasets has
been analyzed using statistical method ANOVA. ANOVA being a powerful statistical tool compares the mean of
datasets and determines that if there is a significant difference between them as summarized in Table 3.
Ho (Null hypothesis) = there is no significant difference between the means of datasets being compared.
H1 (alternate hypothesis) = there is a significant difference between the means of datasets being compared.
Results of ANOVA have been listed in Table 4. Total variability of data is calculated by sum of squares (SS).
Degree of freedom represent the number of independent observations available to estimate every response.
F-statistics and associated p-value are significant results obtained from ANOVA test. F-statistics determines the
variability between the groups to the variability within the group. p value presents the probability to observe a
difference as large as the one observed in response set.
The f-ratio value is 100.8232. The p value is < 0.00001. The result is significant at p < 0.05. There is a significant
difference between the means being compared. The p value is less than the commonly used significance level

Figure 3.  Proposed model workflow.

Figure 4.  Experimental setup.

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 5

Vol.:(0123456789)
 www.nature.com/scientificreports/

Classes 1 2 3 4 5 Total
Sample (X) 70 70 70 70 70 350
∑X 245 236 2142 245 236 3104
Mean (k) 3.5 3.3714 30.6 3.5 3.3714 8.869
∑X2 1155 880 100,132 1155 880 104,202
Std.Dev. (σ) 2.0764 1.1056 22.3888 2.0764 1.1056 14.8222

Table 3.  Analysis of variance summary using ANOVA.

Source Sum of Squares (SS = Σσ/Σk) Degree of freedom (df) Mean square (MS = ∑X2/df)
Between-datasets 41,323.47 4 10,330.87
Within-datasets 35,350.49 345 102.4652
Total 76,673.95 349

Table 4.  ANOVA results.

(0.05), it can be inferred that H

­ o has been rejected and can be concluded that H
­ 1 has been accepted indicating
the significant difference between the means being compared.
Second dimension of analysis part focused on visualizing performance of global model trained on central
server through meta classifier by drawn receiver operating characteristic (ROC) curve. Data characteristics of
ROC curve are presented in Table 5. Figure 5 compares the performance of global model on the basis of sensitiv-
ity [TP/(TP + FN)] and specificity [TP/(TP + FN)].

Number of actual negative cases = 31

Number of actual positive cases = 26
Response data
Category 1 2 3 4 5
Actual Negative cases 11 12 5 1 2
Actual positive cases 1 2 4 9 10
Observed operating points
FPF 0.0000 0.645 0.968 0.2581 0.6452 1.0000
TPF 0.0000 0.3846 0.7308 0.8846 0.9615 1.0000
Initial values of parameters Final values of parameters
A = 1.5647 Procedure converges after 6 iterations
B = 0.9591 A = 1.7403
Z(K): − 0.3718 0.6490 1.3003 1.5182 B = 1.1114
Z(K): − 0.3322 0.5665 1.1380 1.7785
LOGL = − 83.3485
LOGL = − 77.6311
Variance–covariance matrix Correlation matrix
A 1.0000 0.6688 0.4215 0.4991 0.3029
A 0.2089 0.1097 0.0440 0.0520 0.0367 − 0.0012
− 0.0070
B 0.6688 1.0000 0.2225 0.0592
B 0.1097 0.1288 0.0182 0.0048 − 0.0285 − 0.0873
− 0.3004 − 0.6424
Z(1) 0.4215 0.2225 1.0000 0.5413
Z(1) 0.0440 0.0182 0.0521 0.0282 0.0200 0.0110
0.3309 0.1272
Z(2) 0.4991 0.0592 0.5413 1.0000
Z(2) 0.0520 0.0048 0.0282 0.0519 0.0425 0.0378
0.7050 0.4388
Z(3) 0.3029 − 0.3004 0.3309 0.7050
Z(3) 0.0367 − 0.0285 0.0200 0.0425 0.0701 0.0766
1.0000 0.7642
Z(4) − 0.0070 − 0.6424 0.1272 0.4388
Z(4) − 0.0012 − 0.0873 0.0110 0.0378 0.0766 0.1432
0.7642 1.0000
Summary of ROC curve
Area = 0.8778
Std. dev. (area) = 0.0461

Table 5.  Data characteristics (ROC curve).

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 6

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 5.  ROC curve.

Validation. In response set, data points have been gathered into one of the following four classes to validate
ASD diagnosis. Class1: true positive (TP) indicates that the person has autism, and we have correctly recorded
autism positivity. Class 2: true negative (TN) means that a person does not has autism and wrongly recorded as
negative in response dataset. Class 3: false positive (FP) depicts that response dataset incorrectly recorded that a
person had ASD who does not have it. Class 4: false negative (FN) indicates that it was predicted mistakenly that
the person does not have ASD, but they have ASD. The confusion matrix of ASD that facilitated in the validation
process is given below in Table 6.
Precision, recall and F1 score are the measures used to validate performance of LR and SVM classifiers. Preci-
sion demonstrates the cases that detected autism and we predicted them correctly. Whereas recall indicates the
number of autism cases identified correctly are relevant out of total instances that had autism. Proposed model
has been validated using dataset B, D given in Table1.
F1 score greater than 0.5 or above is considered Good. It can be observed from Table 7 that SVM is performing
more accurately than LR although LR is also giving comparable results. Hence, it can be inferred from results that
SVM and LR can detect autism more accurately in comparison of other ML models using features dataset and
they can be used for early diagnosis of autism. Figures 6 and 7 present precision and recall curve of SVM and LR
respectively. Precision and recall are the measures used to evaluate model’s performance. Precision demonstrates

Detection Person with ASD Person without ASD

ASD detected TP FP
ASD not detected FN TN

Table 6.  ASD prediction matrix.

Children dataset Adult dataset

Matrices LR SVM SVM LR
Accuracy 98% 99% 81% 78%
Precision 0.92 0.97 0.73 0.81
Recall 0.44 0.58 0.56 0.51
       
24 2 34 1 45 17 30 7
Confusion matrix 2 31 5 25 14 35 13 29

F1 score 0.60 0.73 0.63 0.62

Table 7.  Performance of ML models in ASD detection.

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 7

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 6.  Precision/Recall curve of SVM.

Figure 7.  Precision/Recall curve of LR.

the cases that detected autism and we predicted them correctly. Whereas recall indicates how many autism cases
model has identified correctly as relevant out of total instances that had autism.
After performing detailed analysis, it has been observed that SVM and LR models can be best fit for diagno-
ses of autism disorder in people of various age groups ranging from children to adults. We have obtained 99%
accuracy in prediction of ASD.
The performance of proposed model has also been compared with other models already proposed in the
literature. We found three most relevant studies that have proposed models for ASD detection.

Ethical statement. Hereby, I Muhammad Shoaib Farooq consciously assure that for the manuscript
“Detection of Autism Spectrum Disorder (ASD) in children and adults using Machine Learning” the following is
fulfilled: (1) This material is the authors’ own original work, which has not been previously published elsewhere.
(2) The paper is not currently being considered for publication elsewhere. (3) The paper reflects the authors’ own
research and analysis in a truthful and complete manner. (4) The paper properly credits the meaningful contri-
butions of co-authors. (5) The results are appropriately placed in the context of prior and existing research. (6)
All sources used are properly disclosed (correct citation). Literally copying of text must be indicated as such by
using quotation marks and giving proper reference. (7) All authors have been personally and actively involved
in substantial work leading to the paper, and will take public responsibility for its content. The violation of the
Ethical Statement rules may result in severe consequences. I agree with the above statements and declare that this
submission follows the policies as outlined in the Guide for Authors and in the Ethical Statement.

Discussion
Table 2 indicates the response set gathered by analysing multiple features extracted during pre-processing of
datasets. Figures 8 and 9 have been drawn based upon response R1 that showed the region to which most of ASD
patients belong and their ethnicity. It can be observed from the chart that United Kingdom (UK) is the most
affected region. Similarly, graph in Fig. 9 presents that mostly White-Europeans have ASD.
People infected with jaundice (response R8) are considered as on high risk of ASD. So, it is worthwhile to
know that whether a person is born with or without jaundice. There is a high probability that they will screen
positive for ASD if born with jaundice as shown in Fig. 10.

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 8

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 8.  ASD detection as per country-of-residence.

Figure 9.  ASD detection as per ethnicity.

Figure 10.  ASD detection based on jaundice.

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 9

Vol.:(0123456789)
 www.nature.com/scientificreports/

Application of ML in autism detection has significance due to its reliability, accuracy and q ­ uickness1. In the
proposed model, datasets have been processed to train LR and SVM classifiers locally. Results of these classifiers
are transmitted to central server where meta classifier is trained to generate global model for autism detection.
The reason for selecting LR is to find a model that most accurately describes the relationship among binary
response set and independent variables s­ et5. SVMs has been applied in this study as datasets had multiple dimen-
sions and are not linearly separable. SVM use hyperplane that separates ASD dataset into two classes namely
ASD effected and Non-ASD to predict target and handle overfitting as well. SVM has separating hyper plane
boundary to separate both c­ lasses7 as presented in Fig. 11.

Comparison with other studies. We have compared their work with our proposed model and summa-
rized the strengths and limitations of existing model in relation to our proposed models in Table 8. It has been
noted that our proposed model is offering comparable accuracy and effectively applicable to diagnose ASD in
patients belonging to different age groups ranging from children to adults.

Limitations of proposed model. FL is a ML technique that allows models to be trained on decentralized

data sources without transferring the data to a central server. Proposed FL based model for ASD detection offers
several advantages of data security and data privacy but it has some limitations too as listed below:
Limited model complexity In proposed architecture, FL models are trained on multiple devices with limited
processing power and storage. This limitation can make it difficult to use the proposed model for more complex
tasks that require deep neural networks or other advanced machine learning models.
Data heterogeneity The proposed model is designed to work with data that is distributed across different
devices and locations. However, this can lead to data heterogeneity, where different devices have different types
of data, making it challenging to develop models that perform well across all devices.

Figure 11.  SVM mechanism of ASD Classification.

Refs Dataset Models Accuracy Limitations Strengths

LR : 93.15% The dataset was compiled from
NB:94.79% primarily autism-based collec- Used forward feature selection
1 Children 1054 instances along
LR, SVM, KNN, RF SVM:93.84% tions, as a result of which there Trained and tested five ML
with 18 attributes
KNN:90.52% was quite a significant imbal- models
RFC: 81.52% ance, in favor of the ASD class
5 2009 features records (toddlers, SVM, Glmboost and adaboost Constrained sample size/ Metrics based on brain activity
85.10%, 97%, 98%
children, adults) classification data set used for prediction of ASD
Does not predict the severity
of ASD. Conditions used for Makes use of six different ML
Naïve Bayes, SVM, LR, RF,
7
Children have 292 instances 97.53%, 9 6.30%, 96.88% identification of ASD which based classification methods
CNN, NN
might not always necessarily and obtained high accuracy
translate to a case of ASD
Comparable accuracy has been
Children Children obtained
Records: > 200 SVM: 99% Two different datasets of
Attributes: 22 LR: 98% Size of children dataset is very children and adults have been
Proposed models SVM, LR
Adults Adults small processed
Records: > 700 SVM: 81% State of the art FL technique
Attributes: 21 LR: 78% has been applied for ASD
prediction

Table 8.  A comparison of our proposed models with previous work.

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 10

Vol:.(1234567890)
www.nature.com/scientificreports/

Communication overhead In the proposed architecture, models are trained on local devices, and the updated
models need to be sent back to a central server for aggregation. This process can create significant communication
overhead, especially when dealing with a large number of devices or when the models are updated frequently.
Lack of transparency The proposed model for ASD detection, makes it challenging to understand how models
are trained or how they make predictions. This lack of transparency can make it difficult to identify and correct
biases or errors in the models.

Conclusion
The assessment of ASD has been associated with multiple disorders recognized as features including, behavioural,
emotional, structural and mental disorders that make it difficult to predict due to non-availability of medical
tests for all features needed to detect ASD in a person. Practitioners diagnose ASD in patients by using psycho-
logical assessments and response observation. Detection process is time-consuming and complex as symptoms
are not obvious. Presently, there is no screening method that has been optimized and thoroughly developed
to specifically detect the ASD, nor is there a screening test that can accurately diagnose ASD. ML is the most
recent development that can facilitate in predicting autism more accurately saving lots of time. ML can be help-
ful in early diagnosis of ASD in patients of all ages including children and adults. In this work, we have applied
two different ML models (SVM, LR) on the dataset containing features of children and adults. It was observed
that SVM showed 81% accuracy in detecting ASD in adults and LR gave 98% accuracy in determining ASD in
children. In future, different transfer-learning models i.e. MobileNet, ResNet can also be used in ASD detection
using images dataset of autistic children for early detection of ASD with improved accuracy. Moreover, severity
of disorder can also be measured through deep learning methods in future.

Data availability
Autism image dataset for children : Cihan063, https://​www.​kaggle.​com/​datas​ets/​cihan​063/​autism-​image-​data,
Accessed on: 05 June 2022. Autism Screening on Adults : and rewmvd, https://​www.​kaggle.​com/​datas​ets/​andre​
wmvd/​autism-​scree​ning-​on-​adults, Accessed on: 05 June 2022.

Received: 7 October 2022; Accepted: 25 May 2023

References
1. Vakadkar, K., Purkayastha, D. & Krishnan, D. Detection of autism spectrum disorder in children using machine learning tech-
niques. SN Comput. Sci. 2(5), 1–9 (2021).
2. Park, M. N., Moulton, E. E. & Laugeson, E. A. Parent-assisted social skills training for children with autism spectrum disorder:
PEERS for preschoolers. Focus Autism Dev. Disabil. https://​doi.​org/​10.​1177/​10883​57622​11101​58 (2022).
3. Gosling, C. J. et al. Efficacy of psychosocial interventions for Autism spectrum disorder: An umbrella review. Mol. Psychiatry 27,
1–10 (2022).
4. Willsey, H. R., Willsey, A. J., Wang, B. & State, M. W. Genomics, convergent neuroscience and progress in understanding autism
spectrum disorder. Nat. Rev. Neurosci. 23(6), 323–341 (2022).
5. Rahman, M. M. et al. A Review of machine learning methods of feature selection and classification for autism spectrum disorder.
Brain Sci. 10(12), 949 (2020).
6. Akter, T. et al. Machine learning-based models for early stage detection of autism spectrum disorders. IEEE Access 7, 166509–
166527 (2019).
7. Wei, Q., Xu, X., Xu, X. & Cheng, Q. Early identification of autism spectrum disorder by multi-instrument fusion: A clinically
applicable machine learning approach. Psychiatry Res. 320, 115050 (2023).
8. Yaneva, V., Eraslan, S., Yesilada, Y. & Mitkov, R. Detecting high-functioning autism in adults using eye tracking and machine
learning. IEEE Trans. Neural Syst. Rehabil. Eng. 28(6), 1254–1261 (2020).
9. Jamwal, I., Malhotra, D. & Mengi, M. A systematic study of intelligent autism spectrum disorder detector. Int. J. Comput. Vis.
Robot. 13(2), 219–234 (2023).
10. Hosseinzadeh, M. et al. A review on diagnostic autism spectrum disorder approaches based on the Internet of Things and machine
learning. J. Supercomput. 77(3), 2590–2608 (2021).
11. Eslami, T. & Saeed, F. Auto-ASD-network: A technique based on deep learning and support vector machines for diagnosing autism
spectrum disorder using fMRI data. In Proceedings of the 10th ACM International Conference on Bioinformatics, Computational
Biology and Health Informatics, 646–651 (2019).
12. Yuan, L., Erdt, M., Li, R. & Siyal, M. Y. Data privacy protection domain adaptation by roughing and finishing stage. Vis. Comput.
https://​doi.​org/​10.​1007/​s00371-​023-​02794-1 (2023).
13. Erforth, B. & Martin-Shields, C. Where privacy meets politics: EU–Kenya cooperation in data protection. In Africa–Europe Coop-
eration and Digital Transformation, 142–155 (Routledge, 2023).
14. Zhu, J., Cao, J., Saxena, D., Jiang, S. & Ferradi, H. Blockchain-empowered federated learning: Challenges, solutions, and future
directions. ACM Comput. Surv. 55(11), 1–31 (2023).
15. Tehseen, R., Farooq, M. S. & Abid, A. A framework for the prediction of earthquake using federated learning. PeerJ Comput. Sci.
7, e540 (2021).
16. Farooq, M. S. et al. FFM: Flood forecasting model using federated learning. IEEE Access 11, 24472–24483 (2023).
17. Nigmatullina, I., Sheymardanov, S. & Abramskiy, M. Digital platform for monitoring and comprehensive support of children with
autism spectrum disorders. In Intelligent Sustainable Systems: Selected Papers of WorldS4 2022, vol. 1, 573–580 (Springer Nature
Singapore, 2023).
18. Ali, S. et al. A multi-centre polyp detection and segmentation dataset for generalisability assessment. Sci. Data 10(1), 75 (2023).
19. Ghosh, T., Banna, M. H. A., Nahian, M. J. A., Kaiser, M. S., Mahmud, M., Li, S. & Pillay, N. A privacy-preserving federated-
mobilenet for facial expression detection from images. In Applied Intelligence and Informatics: Second International Conference,
AII 2022, Reggio Calabria, Italy, September 1–3, 2022, Proceedings, 277–292. (Springer, 2023).
20. Francés, L. et al. An approach for prevention planning based on the prevalence and comorbidity of neurodevelopmental disorders
in 6-year-old children receiving primary care consultations on the island of Menorca. BMC Pediatr. 23(1), 1–14 (2023).
21. Alfalasi, M. M. B. B. Early detection of autism spectrum disorder (ASD) using machine learning techniques (2023).
22. Cao, X. & Cao, J. Commentary: Machine learning for autism spectrum disorder diagnosis–challenges and opportunities–a com-
mentary on Schulte-Rüther et al. (2022). J. Child Psychol. Psychiatry 64, 966–967 (2023).

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 11

Vol.:(0123456789)
 www.nature.com/scientificreports/

23. Zhu, F. et al. Multi-modal machine learning system in early screening for toddlers with autism spectrum disorders based on
response to name. Front. Psychiatry 14, 34 (2023).
24. Elbattah, M., Carette, R., Cilia, F., Guérin, J. L. & Dequen, G. Applications of machine learning methods to assist the diagnosis of
autism spectrum disorder. In Neural Engineering Techniques for Autism Spectrum Disorder, vol. 2, 99–119 (Academic Press, 2023).
25. Lawan, A. A., Cavus, N., Abdulrazak, U. I. & Tahir, S. Fundamentals of machine-learning modeling for behavioral screening and
diagnosis of autism spectrum disorder. In Neural Engineering Techniques for Autism Spectrum Disorder, vol. 2, 253–268 (Academic
Press 2023).
26. Cantin-Garside, K. D. et al. Detecting and classifying self-injurious behavior in autism spectrum disorder using machine learning
techniques. J. Autism Dev. Disord. 50(11), 4039–4052 (2020).
27. Beary, M., Hadsell, A., Messersmith, R. & Hosseini, M. P. Diagnosis of autism in children using facial analysis and deep learning.
arXiv preprint https://​arxiv.​org/​abs/​2008.​02890 (2020).
28. Derbali, M., Jarrah, M. & Randhawa, P. Autism spectrum disorder detection: Video games based facial expression diagnosis using
deep learning. Int. J. Adv. Comput. Sci. Appl. 14(1), 110–119 (2023).
29. Devika, K., Mahapatra, D., Subramanian, R. & Oruganti, V. R. M. Outlier-based autism detection using longitudinal structural
MRI. IEEE Access 10, 27794–27808 (2022).
30. Makhnytkina, O., Frolova, O. & Lyakso, E. Morphological and emotional features of the speech in children with typical develop-
ment, autism spectrum disorders and down syndrome. In Artificial Intelligence and Natural Language: 11th Conference, AINL 2022,
Saint Petersburg, Russia, April 14–15, 2022, Revised Selected Papers, 49–59 (Springer, 2023).
31. Liu, R. et al. Spatial–temporal co-attention learning for diagnosis of mental disorders from resting-state fMRI data. IEEE Trans.
Neural Netw. Learn. Syst. https://​doi.​org/​10.​1109/​TNNLS.​2023.​32430​00 (2023).
32. Lord, C., Elsabbagh, M., Baird, G. & Veenstra-Vanderweele, J. Autism spectrum disorder. Lancet 392(10146), 508–520 (2018).
33. Bilic, P. et al. The liver tumor segmentation benchmark (lits). Med. Image Anal. 84, 102680 (2023).
34. Husna, R. N. S., Syafeeza, A. R., Hamid, N. A., Wong, Y. C. & Raihan, R. A. Functional magnetic resonance imaging for autism
spectrum disorder detection using deep learning. J. Teknol. 83(3), 45–52 (2021).
35. Liu, Q., Dou, Q., Chen, C. & Heng, P. A. Domain generalization of deep networks for medical image segmentation via meta learn-
ing. In Meta-learning with Medical Imaging and Health Informatics Applications, 117–139 (Academic Press, 2023).
36. Nogay, H. S. & Adeli, H. Machine learning (ML) for the diagnosis of autism spectrum disorder (ASD) using brain imaging. Rev.
Neurosci. 31(8), 825–841 (2020).
37. Subah, F. Z., Deb, K., Dhar, P. K. & Koshiba, T. A deep learning approach to predict autism spectrum disorder using multisite
resting-state fMRI. Appl. Sci. 11(8), 3636 (2021).
38. Xu, L. et al. Identification of autism spectrum disorder based on short-term spontaneous hemodynamic fluctuations using deep
learning in a multi-layer neural network. Clin. Neurophysiol. 132(2), 457–468 (2021).
39. Yin, W., Mostafa, S. & Wu, F. X. Diagnosis of autism spectrum disorder based on functional brain networks with deep learning. J.
Comput. Biol. 28(2), 146–165 (2021).
40. Shenouda, J. et al. Prevalence and disparities in the detection of autism without intellectual disability. Pediatrics 151(2), e2022056594
(2023).
41. Wawer, A., Chojnicka, I., Okruszek, L. & Sarzynska-Wawer, J. Single and cross-disorder detection for autism and schizophrenia.
Cogn. Comput. 14(1), 461–473 (2022).
42. Alhassan, S., Soudani, A. & Almusallam, M. Energy-efficient EEG-based scheme for autism spectrum disorder detection using
wearable sensors. Sensors 23(4), 2228 (2023).
43. Ali, N. A., Syafeeza, A. R., Jaafar, A. S., Alif, M. K. M. F. & Ali, N. A. Autism spectrum disorder classification on electroencepha-
logram signal using deep learning algorithm. IAES Int. J. Artif. Intell. 9(1), 91–99 (2020).
44. Sujana, D. S. & Augustine, D. P. Diagnosis of autism spectrum disorder: A review of three focused interventions. SN Comput. Sci.
4(2), 139 (2023).
45. ElNakieb, Y. et al. Understanding the role of connectivity dynamics of resting-state functional MRI in the diagnosis of autism
spectrum disorder: A comprehensive study. Bioengineering 10(1), 56 (2023).
46. Niu, K. et al. Multichannel deep attention neural networks for the classification of autism spectrum disorder using neuroimaging
and personal characteristic data. Complexity https://​doi.​org/​10.​1155/​2020/​13578​53 (2020).
47. Reza, S. M. et al. Deep-learning-based whole-lung and lung-lesion quantification despite inconsistent ground truth: Application
to computerized tomography in SARS-CoV-2 nonhuman primate models. Acad. Radiol. https://​doi.​org/​10.​1016/j.​acra.​2023.​02.​
027 (2023).
48. Singh, A. et al. Machine learning in autism spectrum disorder diagnosis and treatment: Techniques and applications. Neural Eng.
Tech. Autism Spect. Disord. 2, 173–193 (2023).
49. Jacob, S. G., Sulaiman, M. M. B. A. & Bennet, B. Feature signature discovery for autism detection: An automated machine learning
based feature ranking framework. Comput. Intell. Neurosci. https://​doi.​org/​10.​1155/​2023/​63300​02 (2023).
50. Ahmed, I. A. et al. Eye tracking-based diagnosis and early detection of autism spectrum disorder using machine learning and deep
learning techniques. Electronics 11(4), 530 (2022).
51. Rabbi, M. F., Zohra, F. T., Hossain, F., Akhi, N. N., Khan, S., Mahbub, K. & Biswas, M. Autism spectrum disorder detection using
transfer learning with VGG 19, inception V3 and DenseNet 201. In Recent Trends in Image Processing and Pattern Recognition:
5th International Conference, RTIP2R 2022, Kingsville, TX, USA, December 1–2, 2022, Revised Selected Papers, 190–204 (Springer,
2023).
52. Raj, S. & Masood, S. Analysis and detection of autism spectrum disorder using machine learning techniques. Procedia Comput.
Sci. 167, 994–1004 (2020).
53. Ullah, F. et al. Fusion-based body-worn IoT sensor platform for gesture recognition of autism spectrum disorder children. Sensors
23(3), 1672 (2023).
54. Tehseen, R., Farooq, M. S. & Abid, A. EPS: An earthquake prediction system using federated learning. In 2021 International
Conference on Innovative Computing (ICIC), 1–8. (IEEE, 2021).
55. Chaddad, A., Peng, J., Xu, J. & Bouridane, A. Survey of explainable AI techniques in healthcare. Sensors 23(2), 634 (2023).
56. Sundas, A., Badotra, S., Rani, S. & Gyaang, R. Evaluation of autism spectrum disorder based on the healthcare by using artificial
intelligence strategies. J. Sens. https://​doi.​org/​10.​1155/​2023/​53823​75 (2023).
57. Kaur, P. & Kaur, A. Review of progress in diagnostic studies of autism spectrum disorder using neuroimaging. Interdiscip. Sci.
Comput. Life Sci. 15, 1–20 (2023).
58. Voinsky, I., Fridland, O. Y., Aran, A., Frye, R. E. & Gurwitz, D. Machine learning-based blood RNA signature for diagnosis of
autism spectrum disorder. Int. J. Mol. Sci. 24(3), 2082 (2023).

Author contributions
M.S.F. and R.T. performed the measurements and analysis of the article. R.T., M.S.F. were involved in planning
and supervised the research work. Z.A., R.T. and M.S.F. processed the experimental data, performed the analysis,
drafted the manuscript and designed the figures. M.S.F. and M.S. obtained the dataset and characterized it. R.T.
and M.S.F. performed the experimental work and worked on different analysis tools and article repositories.

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 12

Vol:.(1234567890)
www.nature.com/scientificreports/

M.S.F., Z.A., and R.T. aided in interpreting the results and worked on drafting the manuscript. All authors dis-
cussed the results and commented on the whole manuscript.

Funding
No Funding Available for this research.

Competing interests
The authors declare no competing interests.

Additional information
Correspondence and requests for materials should be addressed to Z.A.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.

© The Author(s) 2023

Scientific Reports | (2023) 13:9605 | https://doi.org/10.1038/s41598-023-35910-1 13

Vol.:(0123456789)
 © The Author(s) 2023. This work is published under
http://creativecommons.org/licenses/by/4.0/(the “License”). Notwithstanding
the ProQuest Terms and Conditions, you may use this content in accordance
with the terms of the License.