Iangola Andrianarison
Computer Science MSc student at University of Montreal. Completed my BSc in neuroscience and I am interested in using machine learning working with neuroscience data.
Can the integration of fMRI data and machine learning techniques make diagnosis predictions for ADHD?
Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder characterized by persistent patterns of inattention, hyperactivity, and impulsivity that interfere with daily functioning and development. The use of Functional Magnetic Resonance Imaging (fMRI) has become increasingly valuable in understanding the neural correlates of ADHD and exploring potential diagnostic markers. Machine learning techniques have been applied to analyze fMRI data for diagnosing ADHD and predicting treatment outcomes. Classification models can leverage fMRI features to distinguish between individuals with ADHD and healthy controls, as well as to predict clinical outcomes or treatment responses. This project uses machine learning techniques to train models to make prediction of ADHD diagnosis.
- Become more familiar with classification machine learning models and working with Scikit Learn
- Become more familiar with fMRI and how to work with fMRI data (preprocessing, visualization, connectivity analysis)
- Develop good project management practices
The data used in this project is the Nitrc ADHD resting-state dataset. The preprocessed data is available on Nilearn for 40 participants.
- Git and GitHub for version control
- fMRI data visualization, connectivity analysis with Nilearn
- plotly.express and ipywidgets to create interactive visualizations
- Machine learning with scikit learn with Python
- Jupyter notebook containing the code, and the visualization
- A Github repository for the project
- Jupyter notebook containing the code, visualizations, performances analysis and comparisons