Dual Output Machine Learning Pipeline for Mental Attention Disorder and Gender Prediction Using fMRI and Behavioral Data

DOI

10.2991/978-94-6463-978-0_54How to use a DOI?

Keywords

ADHD; AI diagnosis; XGBoost; LightGBM; fMRI; Hyperparameter Tuning; Dual-output prediction

Abstract

This research focuses on children affected by ADHD, aiming to decode patterns using clinical, behavioral, and fMRI data. The team used ML and DL models such as multilayer perceptron, XGBoost, and LightGBM and then took the best instances of each model to then ensemble them into a hybrid ML model. This proposed model introduces a ML pipeline that predicts the gender of the patient as the diagnosis for ADHD in women is much harder compared to finding the symptoms in men. Unlike previously incorporated uni-model approaches, this framework focuses on both the diagnostic accuracy and gender-specific patterns for further insight into treatment. Among various models, LightGBM delivered the best results with an accuracy of 79.8%. Detailed metrics include a precision of 91.43, recall of 96.34, RMSE of 86.69, and F1 score of 88.64. The accuracy is lower compared to already published other works is due to vast amount of behavioural data the team has incorporated including the fMRI and clinical data. Most studies does not include the behavioural data into these studies.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial 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 license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Mohit Yadav
AU  - Anjali Pujari
AU  - Suhani Alatkar
AU  - Aditi Agasgekar
AU  - Niranjan Muchandi
AU  - Salma Shahapur
PY  - 2025
DA  - 2025/12/31
TI  - Dual Output Machine Learning Pipeline for Mental Attention Disorder and Gender Prediction Using fMRI and Behavioral Data
BT  - Proceedings of the 1st Engineering Data Analytics and Management Conference (EAMCON 2025)
PB  - Atlantis Press
SP  - 636
EP  - 648
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-978-0_54
DO  - 10.2991/978-94-6463-978-0_54
ID  - Yadav2025
ER  -