ORCID
- Kush Gupta: 0009-0008-9930-6435
- Amir Aly: 0000-0001-5169-0679
- Emmanuel Ifeachor: 0000-0001-8362-6292
- Rohit Shankar: 0000-0002-1183-6933
Document Type
Conference Proceeding
Abstract
Autism spectrum disorder (ASD) is a neurodevelopmentalcondition characterized by atypical brain maturation. However, the adaptationof transfer learning paradigms in machine learning for ASD researchremains notably limited. In this study, we propose a computeraideddiagnostic framework with two modules. This chapter presents atwo-module framework combining deep learning and explainable AI forASD diagnosis. The first module leverages a deep learning model finetunedthrough cross-domain transfer learning for ASD classification. Thesecond module focuses on interpreting the model’s decisions and identifyingcritical brain regions. To achieve this, we employed three explainableAI (XAI) techniques: saliency mapping, Gradient-weighted Class ActivationMapping, and SHapley Additive exPlanations (SHAP) analysis.This framework demonstrates that cross-domain transfer learning caneffectively address data scarcity in ASD research. In addition, by applyingthree established explainability techniques, the approach reveals howthe model makes diagnostic decisions and identifies brain regions mostassociated with ASD. These findings were compared against establishedneurobiological evidence, highlighting strong alignment and reinforcingthe clinical relevance of the proposed approach.
DOI Link
Publication Date
2026-01-01
Event
BIOSTEC: 18th International Joint Conference on Biomedical Engineering Systems and Technologies: HEALTHINF: 18th International on Health Informatics
Publication Title
Biomedical Engineering Systems and Technologies - 18th International Joint Conference, BIOSTEC 20255, Revised Selected Papers
Volume
3115
Publisher
Springer Nature
ISBN
978-3-032-34458-8, 978-3-032-34459-5
ISSN
1865-0929
Acceptance Date
2025-01-01
Deposit Date
2025-09-17
Funding
We want to thank EPSRC DTP HMT for funding this project. Also, this manuscript was prepared using a limited-access dataset obtained from the Child Mind Institute Biobank, HBN dataset. This manuscript reflects the views of the authors and does not necessarily reflect the opinions or views of the Child Mind Institute.
Additional Links
Keywords
Cross-Domain Transfer Learning, Explainable AI, Saliency Maps, Grad-CAM, SHAP
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
First Page
87
Last Page
110
Recommended Citation
Gupta, K., Aly, A., Ifeachor, E., & Shankar, R. (2026) 'From Predictions to Explanations: Explainable AI for Autism Diagnosis and Identification of Critical Brain Regions: 18th International Joint Conference, BIOSTEC 2025', Biomedical Engineering Systems and Technologies - 18th International Joint Conference, BIOSTEC 20255, Revised Selected Papers, 3115, pp. 87-110. Springer Nature: Available at: 10.1007/978-3-032-34459-5_5
