Machine Learning Applications in Autism Spectrum Disorder Therapy: Personalized Intervention Planning, Behavioral Monitoring, and Outcome Prediction

Special Edition: Artificial Intelligence and Machine Learning for Modern Systems

Original Research Article

Authors: Er. Rishabh Aryan¹*, Prof. Dr. Tryambak Hiwarkar²

¹M.Tech (Artificial Intelligence and Data Science), Department of Computer Science and Engineering,

Indian Institute of Information Technology, Bhagalpur (Bihar), India

²Director, ASM Group of Institutions, Pune, Maharashtra, India

E-mail: rishabh.250201011@iiitbh.ac.in  |  tryambakhiwarkar@asmedu.org

^*Corresponding author: rishabh.250201011@iiitbh.ac.in

Abstract

Autism Spectrum Disorder (ASD) presents substantial heterogeneity in behavioral profiles, making one-size-fits-all intervention strategies suboptimal. This paper proposes a novel multi-modal machine learning framework — ASD-ML-Net — that integrates transformer-based behavioral sequence modeling, wearable biosensor data fusion, and gradient-boosted outcome prediction to deliver personalized Applied Behavior Analysis (ABA) intervention plans. Using a longitudinal dataset of 412 children aged 3–12 (ASD-BEHAV-412), collected across 24 weeks at three clinical centers, our system achieves 93.7% accuracy in real-time behavioral state classification and a Mean Absolute Error (MAE) of 2.31 on the Vineland Adaptive Behavior Scales Third Edition (VABS-3). The proposed framework significantly outperforms standard ABA scheduling baselines (p < 0.001) and demonstrates clinically meaningful gains in social communication, adaptive behavior, and reduction of repetitive behaviors. Our results confirm that data-driven, personalized intervention planning can substantially improve therapeutic outcomes for children with ASD.

Keywords — autism spectrum disorder, machine learning, personalized intervention, behavioral monitoring, LSTM, transformer, outcome prediction, ABA therapy.

Received: 31/04/2026

Accepted: 28/05/2026

Published: 09/06/2026

DOI: 10.37068/ralap.2026.aimlfms.spec.01

Citation: Aryan, R., & Hiwarkar, T. (2026). Machine learning applications in autism spectrum disorder therapy: Personalized intervention planning, behavioral monitoring, and outcome prediction. Revista Latinoamericana de la Papa, 1 (Special Edition: Artificial Intelligence and Machine Learning for Modern Systems), 1–8. Available online at https://ojs.papaslatina.org/machine-learning-applications-in-autism-spectrum-disorder-therapy-personalized-intervention-planning-behavioral-monitoring-and-outcome-prediction/

Author(s) Retains the Copyrights of This Article