Abstract
This study explored various feature extraction methods for use in automated diagnosis of Attention-Deficit Hyperactivity Disorder (ADHD) from functional Magnetic Resonance Image (fMRI) data. Each participant's data consisted of a resting state fMRI scan as well as phenotypic data (age, gender, handedness, IQ, and site of scanning) from the ADHD-200 dataset. We used machine learning techniques to produce support vector machine (SVM) classifiers that attempted to differentiate between (1) all ADHD patients vs. healthy controls and (2) ADHD combined (ADHD-c) type vs. ADHD inattentive (ADHD-i) type vs. controls. In different tests, we used only the phenotypic data, only the imaging data, or else both the phenotypic and imaging data. For feature extraction on fMRI data, we tested the Fast Fourier Transform (FFT), different variants of Principal Component Analysis (PCA), and combinations of FFT and PCA. PCA variants included PCA over time (PCA-t), PCA over space and time (PCA-st), and kernelized PCA (kPCA-st). Baseline chance accuracy was 64.2% produced by guessing healthy control (the majority class) for all participants. Using only phenotypic data produced 72.9% accuracy on two class diagnosis and 66.8% on three class diagnosis. Diagnosis using only imaging data did not perform as well as phenotypic-only approaches. Using both phenotypic and imaging data with combined FFT and kPCA-st feature extraction yielded accuracies of 76.0% on two class diagnosis and 68.6% on three class diagnosis—better than phenotypic-only approaches. Our results demonstrate the potential of using FFT and kPCA-st with resting-state fMRI data as well as phenotypic data for automated diagnosis of ADHD. These results are encouraging given known challenges of learning ADHD diagnostic classifiers using the ADHD-200 dataset (see Brown et al., 2012).
Highlights
Over the past decade, many researchers have applied statistical analysis to functional Magnetic Resonance Images in order to better understand neuropsychiatric phenomena
We investigated whether reducing functional Magnetic Resonance Image (fMRI) dimensionality over both the spatial and temporal dimensions would improve the diagnostic system’s discrimination of Attention-Deficit Hyperactivity Disorder (ADHD) patients from healthy controls
Using only the reduced imaging features from kPCA-st produced an accuracy of 70.3%, which was statistically better than the baseline, Principal Component Analysis (PCA)-t, and PCA over space and time (PCA-st) (p = 0.0024, p = 0.014, and p = 0.022), but it was not statistically better than the accuracy achieved using only the phenotypic data
Summary
Many researchers have applied statistical analysis to functional Magnetic Resonance Images (fMRIs) in order to better understand neuropsychiatric phenomena. Much of this research has used fMRI to identify group differences between subjects that have a specific neuropsychiatric disorder and healthy controls (Purdon et al, 2011). Each fMRI volume contains roughly ∼ 105 voxel locations, each with a waveform that may be composed of hundreds of time points. Some patterns in these voxel waveforms may be diagnostic for a neuropsychiatric disorder, but there is substantial variance in the data that is not related to diagnosis. The accuracy of computerized diagnosis can be facilitated by feature extraction during preprocessing.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have