ASD-DiagNet: A Hybrid Learning Approach for Autism Spectrum Disorder Detection Using fMRI Data

This repository contains the implementation of ASD-DiagNet, a hybrid learning approach for detecting Autism Spectrum Disorder (ASD) using functional Magnetic Resonance Imaging (fMRI) data. The algorithm is based on the methodology presented in the research paper by Taban Eslami et al. (2019).

---

📄 Research Article

If you use this repository in your research, please cite the following article:

Taban Eslami, Fahad Saeed, Vahid Mirjalili, Alvis Fong, and Angela Laird (2019).
"ASD-DiagNet: A Hybrid Learning Approach for Detection of Autism Spectrum Disorder Using fMRI Data."
Published in Frontiers in Neuroinformatics, 13:70 (2019)
Read the Full Article

---

🔧 Environment Setup

Hardware Requirements

• A server with a CUDA-enabled GPU (compute capability 3.5 or higher)

Software Requirements

• Python (version 3.5 or higher)
• PyTorch (version 0.4.1)
• CUDA (version 8.0 or higher)
• Jupyter Notebook

---

⚙️ Parameter Configuration

Before running the notebook, please configure the following parameters in the first cell:

• Atlas Name: Select one of the following options: "cc200", "aal", or "dosenbach160"
  Example:

  p_ROI = "cc200"

• Number of Folds for Cross-Validation: Integer value specifying the number of folds
  Example:

  p_fold = 10

• Classification Mode: Specify "whole" or "percenter" for the desired classification strategy
  Example:

  p_mode = "percenter"

• Center Name: Required only for per-center classification
  Example:

  p_center = "Stanford"

• Classification Method: Choose one of the following: "ASD-DiagNet", "rf", or "SVM"
  Example:

  p_Method = "ASD-DiagNet"

• Data Augmentation: Specify True or False to enable or disable data augmentation (applicable only for ASD-DiagNet)
  Example:

  p_augmentation = False

---

🛡️ License

This project is licensed under the MIT License. See the LICENSE file for details.

---

Acknowledgment

We extend our gratitude to the research community and collaborators who contributed to the development of this project.

---

Thank you for your interest in ASD-DiagNet. We welcome feedback and contributions. If you encounter any issues, please submit them via the Issues section.