Can a model predict the genetic profile of an individual based on brain regions volumes? There is growing evidence suggesting that genetic variations formally associated to neurodevelopmental disorders have significant effects on brain structures. In this project, the performance of three classifiers will be compared when predicting the genetic status of individuals from brain region volumes in a highly imbalanced dataset (UK BioBank cohort).
Does functional connectivity between brain regions differ in male and female? If yes then fMRI data can be used to distinguish sex on the basis of the difference in functional connectivity. I applied supervised Machine Learning algorithms on the fMRI data to classify sex.
Machine learning models are often used to analyze fMRI data, whether it be a simple classification or regression problem or something more complex. While the focus of a study is often centered on the model architecture, data preprocessing also plays a vital role in a model’s success. This project will explore the effect that various preprocessing options may have on the prediction performance of a machine learning model for age prediction using resting state fMRI.
ADHD subtypes are a controversial aspect of ADHD literature. Most subtypes classifications are based on behavioral and cognitive data but lack biomarkers. Using a multimodal dataset comprised of EEG data as well as self-reported symptoms and behavioral data, we tried to predict the DSM subtypes of each of our 96 participants. Since ADHD has been noted to present itself differently across sexes, we also tried to predict sex. At-rest eeg data and behavioral data proved to be poor predictors of the DSM subtypes. However, self-reported symptoms were a rich predictor of ADHD subtype. Additionally, predicting sex using EEG data yielded the highest decoding accuracies.
Computational Psychiatry is growing trend that applies machine learning methods to psychological disorders. How well can we predict schizophrenia diagnosis from brain activity? This project uses neuroimaging tools from Nilearn, and machine learning tools from scikit-learn to differentiate patients diagnosed with schizophrenia from healthy controls using resting state fmri data.
Using fMRI Data to Predict Autism Diagnoses with Various Machine Learning Models and Cross-Validation Methods
Is autism associated with a distinct neurofunctional signature? If so, how accurately are we able to predict the diagnosis based on fMRI data? In this project, we set out to compare different machine learning models and cross-validation methods to see how well each one was able to predict autism from resting state fMRI data in the ABIDE dataset.
Are neuropsychiatric disorders extreme cases of connectivity patterns that are found in the overall population? Using personality traits as a measure of individual variation and knowing that neuroticism is especially linked with mental disorders we wanted to see if neuroticism in a healthy population was linked with specific patterns of connectivity that could be compared to those common to neuropsychiatric disorders.
An introduction to brain decoding and comparing the results of the seven different classifier on Haxby dataset
Brain decoding is a neuroscience field that concerned about different types of stimuli from information that has already been encoded and represented in the brain by networks of neurons. My goal for this project is learning the fundamentals of brain decoding. Moreover, I compared the performance of seven different common classification approaches including Naive Bayes, Nearest Neighbours, Neural Networks, Logistic Regression, Support vector machine, Decision tree and finally the Artificial Neural Network on Haxby dataset.
This project is about diffusion magnetic resonance (MR) data processing and analysis. It mainly consists of three parts: brain diffusion MR data preprocessing, diffusion MRI images reconstruction, data visualization and left and right hemispherical preprocessed MR images classification. The whole procedures can be found in this Jupyter Notebook file. Explanations about procedures results and other details are given in it. With reproducibility being a primary concern, this project was completed by using open-source softwares/tools (Python, FSL, DIPYPE…) and dataset (dHCP and PRIME).