Detection of Autism Spectrum Disorder in Children and Adults Using Machine Learning Techniques

Detection of Autism Spectrum Disorder in Children and Adults Using Machine Learning Techniques

IEEE BASE PAPER TITLE:

Automated Autism Assessment With Multimodal Data and Ensemble Learning: A Scalable and Consistent Robot-Enhanced Therapy Framework

IEEE BASE PAPER ABSTRACT:

Navigating the complexities of Autism Spectrum Disorder (ASD) diagnosis and intervention requires a nuanced approach that addresses both the inherent variability in therapeutic practices and the imperative for scalable solutions. This paper presents a transformative Robot-Enhanced Therapy (RET) framework, leveraging an intricate amalgamation of an Adaptive Boosted 3D biomarker approach and Saliency Maps generated through Kernel Density Estimation. By seamlessly integrating these methodologies through majority voting, the framework pioneers a new frontier in automating the assessment of ASD levels and Autism Diagnostic Observation Schedule (ADOS) scores, offering unprecedented precision and efficiency. Drawing upon the rich tapestry of the DREAM Dataset, encompassing data from 61 children, this study meticulously crafts novel features derived from diverse modalities including body skeleton, head movement, and eye gaze data. Our 3D bio-marker approach achieves a remarkable predictive prowess, boasting a staggering 95.59% accuracy and an F1 score of 92.75% for ASD level prediction, alongside an RMSE of 1.78 and an R-squared value of 0.74 for ADOS score prediction. Furthermore, the introduction of a pioneering saliency map generation method, harnessing gaze data, further enhances predictive models, elevating ASD level prediction accuracy to an impressive 97.36%, with a corresponding F1 score of 95.56%. Beyond technical achievements, this study underscores RET’s transformative potential in reshaping ASD intervention paradigms, offering a promising alternative to Standard Human Therapy (SHT) by mitigating therapist variability and providing scalable therapeutic approaches. While acknowledging limitations in the research, such as sample constraints and model generalizability, our findings underscore RET’s capacity to revolutionize ASD management.

PROJECT OUTPUT VIDEO:

ALGORITHM / MODEL USED:

AdaBoost Classifier, Gradient Boosting Classifier.

OUR PROPOSED PROJECT ABSTRACT:

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition that affects social interaction, communication, and behavioral patterns across both children and adults. Early and accurate identification of ASD plays a crucial role in enabling timely intervention, personalized care, and improved quality of life. With the growing availability of behavioral and demographic data, machine learning techniques have emerged as effective tools to support clinical decision-making by identifying patterns that may not be easily detected through traditional assessment methods.

The need for this project arises from the challenges associated with conventional ASD screening processes, which often rely on lengthy questionnaires, expert evaluations, and subjective judgments. These methods can be time-consuming, resource-intensive, and difficult to scale for large populations. An intelligent, data-driven system can assist healthcare professionals, educators, and caregivers by providing a fast, consistent, and reliable preliminary assessment of ASD risk in both children and adults.

In this project, a web-based Autism Spectrum Disorder detection system is developed using Python as the core programming language, with HTML, CSS, and JavaScript for the front end, and Flask as the web framework. The system utilizes a dataset consisting of 996 records with behavioral scores (A1_Score to A10_Score), demographic attributes such as age, gender, ethnicity, country of residence, and additional factors including jaundice history, autism family history, prior app usage, and relational information.

The core of this project relies on implementing two machine learning models: AdaBoost Classifier and Gradient Boosting Classifier are implemented as separate detection modes to classify individuals as ASD or non-ASD. The AdaBoost model achieved 100% accuracy on both training and testing data, while the Gradient Boosting model achieved 100% training accuracy and 97% testing accuracy.

To ensure comprehensive evaluation, the system presents performance metrics such as precision, recall, F1-score, and confusion matrix for each model, along with visual analytics including model accuracy comparison charts and classification distribution charts. The developed system demonstrates the effectiveness of ensemble machine learning techniques in supporting reliable and interpretable ASD detection through an accessible web-based platform.

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS: 

• System : Pentium i3 Processor.
• Hard Disk : 20 GB.
• Monitor : 15’’ LED.
• Input Devices : Keyboard, Mouse.
• Ram : 8 GB.

SOFTWARE REQUIREMENTS: 

• Operating System : Windows 10 / 11.
• Coding Language : Python 3.12.0.
• Web Framework : Flask.
• Frontend : HTML, CSS, JavaScript.

REFERENCE:

Iqbal Hassan, Nazmun Nahid, Minhajul Islam, Shahera Hossain, Björn Schuller, and Md Atiqur Rahman Ahad, “Automated Autism Assessment With Multimodal Data and Ensemble Learning: A Scalable and Consistent Robot-Enhanced Therapy Framework”, IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, VOL. 33, 2025.

👉CLICK HERE TO BUY THIS PROJECT “Detection of Autism Spectrum Disorder in Children and Adults Using Machine Learning Techniques” SOURCE CODE👈

Frequently Asked Questions (FAQ’s) and Answers

Q1. What is the objective of this project?

The objective of this project is to develop a machine learning–based system that can detect Autism Spectrum Disorder (ASD) in children and adults using behavioral and demographic data. The system aims to support early screening and assist decision-making through automated and data-driven analysis.

Q2. Which technologies are used to develop the system?

The system is developed using Python for backend processing, HTML, CSS, and JavaScript for the frontend, and Flask as the web framework. Machine learning models are implemented using standard Python libraries.

Q3. What type of data is used in this project?

The project uses a structured dataset containing 996 records with features such as behavioral screening scores (A1_Score to A10_Score), age, gender, ethnicity, country of residence, medical history indicators, and other demographic attributes related to ASD assessment.

Q4. Which machine learning algorithms are implemented?

Two machine learning algorithms are implemented as separate detection modes: AdaBoost Classifier and Gradient Boosting Classifier. Each model is trained and evaluated independently to classify individuals as ASD or non-ASD.

Q5. How accurate are the machine learning models?

The AdaBoost Classifier achieved 100% accuracy on both training and testing datasets, while the Gradient Boosting Classifier achieved 100% training accuracy and 97% testing accuracy, demonstrating strong predictive performance.

Q6. What evaluation metrics are used to assess model performance?

The system evaluates model performance using precision, recall, F1-score, and confusion matrix. In addition, visual charts such as model accuracy comparison and classification distribution are generated for better analysis.

Q7. Is the system intended to replace clinical diagnosis?

No. The system is designed as a screening and decision-support tool. It assists in identifying potential ASD indicators but does not replace professional clinical diagnosis or expert evaluation.

Q8. Can the system be used for both children and adults?

Yes. The system is designed to support ASD detection across different age groups, making it suitable for both children and adults.

Q9. Does the system require specialized hardware or sensors?

No. The proposed system relies only on questionnaire-based and demographic input data and does not require any specialized hardware, sensors, or robotic devices.

Q10. How user-friendly is the application?

The application provides a simple and intuitive web interface that allows users to enter data easily and view results clearly. It can be used by healthcare professionals, educators, researchers, and caregivers with minimal training.