New ASD Detection Approach by Mennahtullah Mabrouk at JAC-ECC 2025
We are proud to announce that Mennahtullah Mabrouk, one of our exceptional engineers, recently attended the JAC-ECC 2025 Conference where she presented her groundbreaking paper titled "A Hybrid Learning Approach for Detection of Autism Spectrum Disorder Using fMRI Data."
About
Mennahtullah was the sole author of the paper, leading all aspects of the research. Her work introduces an innovative hybrid learning approach for detecting Autism Spectrum Disorder (ASD) using resting-state functional magnetic resonance imaging (fMRI) data. She developed a machine learning pipeline that extracts functional connectivity features from fMRI scans, providing an efficient and interpretable method for classifying ASD that contrasts with traditional deep learning techniques.
In her study, Mennahtullah utilized the ABIDE dataset, incorporating principal component analysis (PCA) for dimensionality reduction, along with a shallow neural network classifier (MLP) and data augmentation techniques. The method achieved a 78.4% classification accuracy, outperforming existing methods and offering a computationally efficient approach for ASD classification.
About Mennahtullah's Work
Mennahtullah's research highlights her ability to balance biological interpretability with computational efficiency, making her approach highly suitable for research settings. The work demonstrates that machine learning can be a powerful tool for interpreting complex brain data, offering promising avenues for future developments in ASD diagnosis. However, it is important to note that while the approach is promising, further research and validation are needed before it can be considered for widespread clinical use.
Her work represents a significant contribution to neuroimaging and machine learning, showcasing how such technologies can be applied to better understand ASD. It paves the way for future advancements in the field, with potential applications in diagnostic tools and research.
We would like to extend our sincere congratulations and deepest thanks to Mennahtullah Mabrouk for her dedication and exceptional work. Her research reflects the core values of Interstellar, and we are incredibly proud to have her as part of our team.
