Predictive Modeling For Children ADHD Detection Through Behavioural Analytics

  • Antony Vigil M S
  • Yaswanth Kumar
  • Sahith Reddy B
  • Sai Rohit S
  • Sai Rohit S
Keywords: ADHD, behavioural activity, children, machine learning, prediction

Abstract

This study presents a novel approach to predictive modelling in the context of behavioural analysis in children by integrating Deep Neural Networks (DNNs) with Transformer-based models. The motivation behind this arises from the fact that behavioral data is complicated and frequently contains non-linear connections and temporal dependencies that traditional models struggle to capture effectively. By combining strengths of DNNs in handling non-linear patterns with the Transformer’s capability to perform sequential By using self-attention processes to process data, the suggested hybrid model seeks to enhance the accuracy and interpretability of predictions. The methodology involved extensive data preprocessing, including normalization, augmentation, and feature engineering to improve the input data's quality. Two distinct architectures were created, with the DNN capturing intricate feature interactions and the Transformer focusing on temporal aspects of the behavioral data. The outputs of both models were fused through a weighted aggregation layer, optimizing predictive performance. Regularization strategies like L2 regularization, batch normalization, and dropout were employed avoid being too oversized and ensure model broad generalization. Hyperparameter tuning was conducted using utilizing Bayesian optimization and grid search leading to the selection of the optimal model configuration. The model’s interpretability was enhanced employing LIME (Local Interpretable Model agnostic Explanations) and SHAP (Shapley Additive explanations), which provided insights into the key behavioral features driving predictions. The results demonstrated that the hybrid model outperformed traditional approaches, with significant improvements in accuracy, precision, and recall. The model’s robustness was confirmed through rigorous assessment with the use of ROC-AUC and F1-Score measures, along with deployment in real-world scenarios where it showed stable performance over time. This study highlights the potential of hybrid models in advancing the field of predictive modeling for behavioral analysis, offering a powerful and interpretable tool for researchers and practitioners working with children’s behavioral data.

Author Biographies

Antony Vigil M S

Department of Computer Science and Engineering, SRM Institute of Science and technology, Ramapuram, Chennai, India

Yaswanth Kumar

Department of Computer Science and Engineering, SRM Institute of Science and technology, Ramapuram, Chennai, India

Sahith Reddy B

Department of Computer Science and Engineering, SRM Institute of Science and technology, Ramapuram, Chennai, India

Sai Rohit S

Department of Computer Science and Engineering, SRM Institute of Science and technology, Ramapuram, Chennai, India

Sai Rohit S

Department of Computer Science and Engineering, SRM Institute of Science and technology, Ramapuram, Chennai, India

References

1. Yina P. Cabarcas-Mena, Andres G. Marrugo, Sonia H. Contreras-Ortiz, “Classification of Cognitive Evoked Potentials for ADHD Detection in Children using Recurrence Plots and CNNs”.
2. Nupur Gupte, Mitali Patel, Tanvi Pen, Swapnali Kurhade , “Early detection of ADHD and Dyslexia from EEG Signals”
3. Sherine M, P. Selvi Rajendran, “Early Detection of ADHD in Children Using Two Learning Approaches”, 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS)
4. Diego Martín, Pablo Casaseca, Susana Alberola, José Antonio López, Francisco Carlos Ruiz, Jesús María Andrés, “Early Detection of ADHD in Juveniles using recurrent neural networks”, 2022 3rd International Conference for Emerging Technology (INCET)
5. Jinhua Ma , Shih-Ya Lin , Xin Chen , Hung-Min Sun, “Automatic diagnosis of ADHD based on nonlinear analysis of actimetry registries”, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
6. Yuan Gao, Xiaotong Wang, Aimin Jiang, Ying Chen, Yibin Tang, “ADHD Diagnosis and Biomarker Detection Based on Multimodal Graph Convolutional Neural Network”, ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
7. Abhilasha Devdatta Kulkarni, Jayashree Rajesh Prasad, “A Comprehensive Literature Review on GestureBased Adaptive Systems for Early Detection of ADHD in Children: Technical Trends and Research Directions”, 2024 MIT Art, Design and Technology School of Computing International Conference (MITADTSoCiCon).
8. Ling He, Fei He, Yuanyuan Li, Xi Xiong, Jing Zhang, “A Robust Movement Quantification Algorithm of Hyperactivity Detection for ADHD Children Based on 3D Depth Images”, IEEE Transactions on Image Processing.
9. Shuanglin Li, Yang Sun, Rajesh Nair, Syed Mohsen Naqvi, “Enhancing ADHD Detection Using Diva Interview-Based Audio Signals and A Two-Stream Network”, 2023 IEEE International Performance, Computing, and Communications Conference (IPCCC).
10. Shuanglin Li, Rajesh Nair, Syed Mohsen Naqvi, "Acoustic and Text Features Analysis for Adult ADHD Screening: A Data-Driven Approach Utilizing DIVA Interview.", IEEE Journal of Translational Engineering in Health and Medicine.
11. Jungpil Shin, Md. Maniruzzaman, Yuta Uchida, Md. Al Mehedi Hasan, Akiko Megumi, Akira Yasumura , “Handwriting-Based ADHD Detection for Children Having ASD Using Machine Learning Approaches” 2023
12. Abhay M S Aradhya, Vigneshwaran Subbaraju, Suresh Sundaram, Narasimhan Sundararajan, "Regularized Spatial Filtering Method (R-SFM) for detection of Attention Deficit Hyperactivity Disorder (ADHD) from resting-state functional Magnetic Resonance Imaging (rs-fMRI)", 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)
13. Ibrahim Traore, Imen Megdiche, Jérôme MarquetDoléac, Lotfi Chaari, "A machine learning technique for device non-wear detection in children with ADHD." , 2023 20th ACS/IEEE International Conference on Computer Systems and Applications (AICCSA).
14. Zarina Begum, Kareemulla Shaik, "A Detailed Examination of Deep Learning Models for Detection of Attention Deficit Hyperactivity Disorder(ADHD)", 2023 6th International Conference on Recent Trends in Advance Computing (ICRTAC)
15. Md. Maniruzzaman, Koki Hirooka, Yoichi Tomioka, Md. Al Mehedi Hasan, Yong Seok Hwang, Akiko Megumi, Akira Yasumura, Jungpil Shin, “Machine Learning-Based ADHD Detection From fNIRs Signal During Reverse Stroop Tasks”, 2024
16. Shashank Jaiswal, Michel F. Valstar, Alinda Gillott, David Daley, “Automatic Detection of ADHD and ASD from Expressive Behaviour in RGBD Data”, 2017 12th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2017)
17. Christian Nash, Rajesh Nair, Syed Mohsen Naqvi , “Machine Learning and ADHD Mental Health Detection - A Short Survey”, 2022 25th International Conference on Information Fusion (FUSION)
18. R.M.I.S Kumarasingha, D.M.C.L.B Dissanayake, P.S.R Pathirathne, K.M.M.Y.S Ranathunga (2023).
“Early Detection and Effective Treatment for ADHD using Machine Learning for Sri Lankan Children” 5th International Conference on Advancements in Computing (ICAC).
19. M.S. Antony Vigil,, Karan Bhavsar, Sanjay Kimbahune, Dr Sundeep Khandelwal, Avik Ghose and Dr Arpan Pal, “Detection of Counterfeit Medicines Using Hyperspectral Sensing”, 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)
20. M.S. Antony Vigil ,Harrish P, Selva J,2022, “Time Series Modelling and Domain specific predicting air passenger flow traffic using Neural Network”.
21. M.S. Antony Vigil ,2018, “A review on rc home automation using Id and IR sensors”, International Journal of Pure and Applied Mathematic”.
Published
2024-10-11
How to Cite
Antony Vigil M S, Yaswanth Kumar, Sahith Reddy B, Sai Rohit S, & Sai Rohit S. (2024). Predictive Modeling For Children ADHD Detection Through Behavioural Analytics. Revista Electronica De Veterinaria, 25(1), 2352-2358. https://doi.org/10.69980/redvet.v25i1.1192
Section
Articles